Display effect processing method and device based on machine learning, equipment and medium
Through a machine learning-based method, using deep learning models to process environment parameters and user behavior, dynamically adjust the display parameters of the display screen, solving the problem that the existing technology cannot adjust the display effect in real time, and achieving the best visual experience in different environments and adaptability.
Patent Information
- Application Number
- CN202410956970.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-08
AI Technical Summary
The existing display technology cannot adjust the display effect dynamically in real time based on the actual usage environment and user posture, resulting in discomfort in brightness and unclear contrast, affecting the user's visual experience.
Using a machine learning-based method, by obtaining real-time environment parameters and user viewing behavior, using preset deep learning models to generate display solutions, dynamically adjust the brightness, color balance, color temperature and gamma curves of the display screen to achieve adaptability to the current environment. display effect.
It provides adaptive display effects in different environments, improves the user's visual experience, reduces the fatigue caused by long-term use of eyes, and adapts to different usage scenarios and user needs.
Smart Images

Figure CN118915931A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of display processing technology, and in particular to a display effect processing method, device, equipment and medium based on machine learning. Background Art
[0002] With the development of display technology and software technology, people have put forward higher requirements for screen display effects, performance and user experience. People have also been working hard to develop new technologies to improve display effects, especially in terms of resolution, but existing display technologies are often unable to dynamically adjust the display effects in real time according to the actual use environment and the user's posture when using it. For example, if the light suddenly increases, but the picture is very dark and the contrast is not clear, users often cannot see the picture clearly, which directly reduces the user's visual experience. These factors that affect the viewing experience include color temperature, brightness, contrast, gamma curve, etc.
[0003] Therefore, in order to provide users with a more comfortable and natural visual experience while reducing eye fatigue caused by prolonged use, how to adaptively and scalably provide users with picture levels and details that match the current environment in different environments is an urgent problem to be solved. Summary of the invention
[0004] Based on this, it is necessary to provide a display effect processing method, device, equipment and medium based on machine learning that can adaptively and scalably provide users with picture levels and details that match the current environment in different environments to address the above technical problems.
[0005] In a first aspect, a display effect processing method based on machine learning is provided, the method comprising:
[0006] Acquire real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, ambient color temperature and change rate;
[0007] Capturing and analyzing the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time;
[0008] Processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation;
[0009] The display screen is controlled to display with the display scheme to obtain a display effect suitable for the current environment.
[0010] In one embodiment, the method further comprises:
[0011] Train the preset deep learning model.
[0012] In one embodiment, the training of a preset deep learning model includes:
[0013] Load pre-trained models and set user preferences;
[0014] Obtain historical environment parameters and historical viewing behaviors;
[0015] Preliminarily screening the historical environmental parameters and historical viewing behaviors through the pre-trained model to remove abnormal data and duplicate data, and supplement missing data through preset probability calculation;
[0016] Normalizing the filtered historical environment parameters and historical viewing behaviors;
[0017] Selecting original features with a high correlation with a target variable from the normalized historical environmental parameters and historical viewing behaviors, wherein the target variable corresponds to a parameter in the display scheme;
[0018] Classify and encode the original features and convert them into numerical features, wherein the classification basis includes user preferences;
[0019] Transforming the original features, wherein the transformation includes logarithmic transformation or square root transformation;
[0020] Reducing the number of the original features by a dimensionality reduction method;
[0021] The regression model is trained using the historical environmental parameters and historical viewing behaviors after dimensionality reduction to obtain the preset deep learning model, and the regression model includes a neural network model or a support vector machine (Support Vector Machine, SVM).
[0022] In one embodiment, the training of the preset deep learning model further includes:
[0023] The preset deep learning model is evaluated using a test data set, wherein the evaluation indicators include one or more combinations of mean square error, root mean square error, and R square.
[0024] In one embodiment, the training of the preset deep learning model further includes:
[0025] The preset deep learning model is optimized according to the evaluation result, and the optimization includes model parameter optimization and model structure optimization.
[0026] In one embodiment, controlling the display screen to display with the display scheme to obtain a display effect adapted to the current environment includes:
[0027] According to the display scheme, combined with the corresponding preset threshold, a real-time display scheme is generated, wherein the real-time display scheme includes one or more combinations of display parameters, adjustment speed, and adjustment mode;
[0028] According to the real-time display scheme, the display parameters of the display screen are dynamically adjusted.
[0029] In one embodiment, after dynamically adjusting the display parameters of the display screen according to the real-time display scheme, the method includes:
[0030] Real-time monitoring of the display screen to adjust the performance of the display effect;
[0031] If the performance is monitored to be poor or the user feedback shows that the current display effect is not good, the display of the display screen is adjusted according to a preset solution.
[0032] In one embodiment, after controlling the display screen to display with the display scheme to obtain a display effect adapted to the current environment, the method further includes:
[0033] Receive feedback from the current user on the display effect corresponding to the display scheme.
[0034] In one embodiment, the receiving of the feedback result of the current user on the display effect corresponding to the display scheme includes:
[0035] Display current display parameters and transition mode to the current user, where the transition mode includes one or more combinations of smooth, fast and intelligent prediction;
[0036] Receive feedback from the current user on the display effect corresponding to the display scheme.
[0037] In one embodiment, the method further comprises:
[0038] According to the feedback results, the preset deep learning model is optimized and updated.
[0039] In a second aspect, a display effect processing device based on machine learning is provided, the device comprising:
[0040] An environmental perception module is used to obtain real-time current environmental parameters, which include any one or a combination of brightness, environmental color temperature and change rate;
[0041] A user perception module is used to capture and analyze the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time;
[0042] A machine learning module, configured to process the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation;
[0043] The display adjustment module is used to control the display screen to display according to the display scheme to obtain a display effect suitable for the current environment.
[0044] In a third aspect, a display device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:
[0045] Acquire real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, ambient color temperature and change rate;
[0046] Capturing and analyzing the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time;
[0047] Processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation;
[0048] The display screen is controlled to display with the display scheme to obtain a display effect suitable for the current environment.
[0049] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:
[0050] Acquire real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, ambient color temperature and change rate;
[0051] Capturing and analyzing the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time;
[0052] Processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation;
[0053] The display screen is controlled to display with the display scheme to obtain a display effect suitable for the current environment.
[0054] The above-mentioned machine learning-based display effect processing method, device, equipment and medium obtain real-time current environmental parameters; capture and analyze the current user's viewing behavior; process the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme; and control the display screen to display with the display scheme to obtain a display effect suitable for the current environment. The present invention can adjust the display parameters of the display screen in combination with the real-time current environmental parameters and viewing behavior, that is, it can achieve the purpose of adjusting the optimal display effect according to different environments and users, and realize that users can still feel the layers and details of the picture in different environments, which not only provides users with a more comfortable and natural visual experience, but also reduces the fatigue caused by long-term eye use, and enables the display device to adapt to different usage scenarios and the needs of different users. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 A schematic diagram of a flow chart of a display effect processing method based on machine learning in one embodiment;
[0056] Figure 2 A schematic diagram of a process for training a preset deep learning model in one embodiment;
[0057] Figure 3 is a flow chart of another display effect processing method based on machine learning in one embodiment;
[0058] Figure 4 is a structural block diagram of a display effect processing device based on machine learning in one embodiment;
[0059] Figure 5 The figure is a diagram showing the internal structure of a device in one embodiment. DETAILED DESCRIPTION
[0060] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0061] In one embodiment, Figure 1 As shown, a display effect processing method based on machine learning is provided, and the method is applied to a display device as an example for explanation. The display device includes an environment perception module, a user perception module, a machine learning module and a display adjustment module, including the following steps:
[0062] Step S11, obtaining real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, environmental color temperature and change rate.
[0063] In an embodiment of the present invention, the real-time current environmental parameters are obtained through the environmental perception module, and the current environmental parameters include any one or a combination of brightness, environmental color temperature and change rate. The environmental perception module is composed of a light sensor and a built-in signal processing module to accurately obtain the light brightness, environmental color temperature and change rate of the current environment, which can suppress environmental noise and has an automatic calibration function to ensure that the output current environmental parameters are reliable.
[0064] Step S12, capturing and analyzing the viewing behavior of the current user, wherein the viewing behavior includes any one or a combination of sitting posture, distance, viewing angle, eye dynamic tracking and viewing time.
[0065] In an embodiment of the present invention, the viewing behavior is data reflecting the user's viewing behavior, and the viewing behavior includes any one or a combination of sitting posture, distance, viewing angle, eye dynamic tracking, and viewing time. The viewing behavior of the current user is captured and analyzed by a user perception module, which is composed of a camera, an infrared sensor, an eye tracking device, etc., and can capture the user's viewing position, viewing angle, line of sight movement and other related parameters in real time. These parameters can be combined with the user's historical usage records and different user preference settings to build a user's behavior model, providing an accurate basis for the model.
[0066] Step S13, processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation.
[0067] In an embodiment of the present invention, the machine learning module processes the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, which includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation. Based on the characteristics of machine learning that can be continuously iterated and optimized, the display effect generated by the optimization is gradually improved, the accuracy and adaptability to different environments are improved, and predictive adjustment can be finally achieved.
[0068] Step S14, controlling the display screen to display with the display scheme to obtain a display effect suitable for the current environment.
[0069] In an embodiment of the present invention, the display adjustment module controls the display screen in real time with the display scheme to obtain a display effect adapted to the current environment, and adjusts the display effect by controlling various display parameters such as the brightness, color balance, color temperature, gamma curve, and natural saturation of the display screen to achieve the best visual effect in the current scene.
[0070] Specifically, step S14 includes
[0071] Sub-step S141, generating a real-time display scheme according to the display scheme and in combination with a corresponding preset threshold, wherein the real-time display scheme includes one or more combinations of display parameters, adjustment speed, and adjustment mode.
[0072] In the embodiment of the present invention, the preset threshold is the optimal threshold of brightness, color balance, color temperature, gamma curve, and natural saturation. The corresponding parameters in the display scheme cannot exceed the corresponding preset threshold. If exceeded, the preset threshold is used as the adjustment value.
[0073] Sub-step S142, dynamically adjusting display parameters of the display screen according to the real-time display solution.
[0074] In an embodiment of the present invention, the display adjustment module dynamically adjusts the display of the display screen to the display parameters by adjusting the speed and the adjustment mode. The adjustment process will be as smooth as possible to avoid sudden changes in the picture quality during the user's viewing process. Specifically, the real-time dynamic adjustment uses a hardware interface or a software API to achieve control of the display screen.
[0075] Further, after the sub-step S142, the method includes:
[0076] Real-time monitoring of the performance of the display screen in adjusting the display effect, including adjustment speed, stability and accuracy;
[0077] If the performance is poor or the user feedback is that the current display effect is poor, the display of the display screen is adjusted according to the preset solution. If the performance is poor or the user feedback is that the current display effect is poor during the display adjustment process, the display of the display screen is adjusted according to the preset solution to correct it in time to avoid affecting the user experience.
[0078] In the above-mentioned display effect processing method based on machine learning, by obtaining real-time current environmental parameters; capturing and analyzing the current user's viewing behavior; processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme; and controlling the display screen to display with the display scheme to obtain a display effect adapted to the current environment, the present invention can adjust the display parameters of the display screen in combination with the real-time current environmental parameters and viewing behavior, that is, it can achieve the purpose of adjusting the optimal display effect according to different environments and users, and realize that users can still feel the layers and details of the picture in different environments, which not only provides users with a more comfortable and natural visual experience, but also reduces the fatigue caused by long-term eye use, and enables the display device to adapt to different usage scenarios and the needs of different users.
[0079] In another embodiment of the present invention, Figure 2 As shown, a step of training a preset deep learning model is provided, including:
[0080] Step S20, load the pre-trained model and set user preferences.
[0081] Step S21, obtaining historical environment parameters and historical viewing behaviors; the historical environment parameters are previous environment parameters, including any one or multiple combinations of brightness, ambient color temperature, and change rate. The historical viewing behaviors are the collected previous user viewing behaviors, including any one or multiple combinations of sitting posture, distance, viewing angle, eye dynamic tracking, and viewing time. The historical environment parameters and historical viewing behaviors are transmitted to the pre-trained model through a specific interface or protocol for processing.
[0082] Step S22, preliminarily screening the historical environmental parameters and historical viewing behaviors through the pre-trained model to remove abnormal data and duplicate data, and supplement missing data through preset probability calculation.
[0083] Step S23, normalize the filtered historical environment parameters and historical viewing behaviors. Since different features may have different dimensions and distributions, it is necessary to normalize the features, which can also be said to be data clipping, so that each feature has a similar scale, which is also a necessary operation for convolutional neural network training.
[0084] Step S24, selecting original features with a high correlation with the target variable from the normalized historical environmental parameters and historical viewing behaviors, where the target variable corresponds to the parameters in the display scheme; according to the purpose of generating the best display effect as needed, features with a high correlation with the target variable, such as ambient light intensity, user viewing distance, viewing time, etc., will be selected for reinforcement learning.
[0085] Step S25, classify and encode the original features and convert them into numerical features, wherein the classification basis includes user preferences; conversion into numerical features facilitates model processing.
[0086] Step S26, transforming the original features, wherein the transformation includes a logarithmic transformation or a square root transformation. Transforming the original features can improve the distribution of features and the performance of the model.
[0087] Step S27, reducing the number of the original features by a dimensionality reduction method, wherein the dimensionality reduction method includes principal component analysis (PCA) and latent Dirichlet Allocation (LDA). Reducing the number of features can reduce the complexity of the model and thus speed up the model's operation speed and resource usage.
[0088] Step S28, use the reduced historical environment parameters and historical viewing behaviors to train the regression model to obtain the preset deep learning model, wherein the regression model includes a neural network model or a support vector machine (SVM). The model is trained with manually labeled historical data in order to train a model of standard data, which is then continuously iterated based on the model. During the training process, cross-validation can also be used to periodically evaluate the performance of the model and select appropriate model parameters. At the same time, regularization, early stopping and other techniques can be used to prevent overfitting.
[0089] Furthermore, after step S28, the method further includes:
[0090] The preset deep learning model is evaluated using a test data set to obtain an evaluation result, wherein the evaluation indicators include one or more combinations of mean-square error (MSE), root mean square error (RMSE), and R-squared.
[0091] Furthermore, after evaluating the preset deep learning model using the test data set and obtaining the evaluation result, the method further includes:
[0092] The preset deep learning model is optimized according to the evaluation result, and the optimization includes model parameter optimization and model structure optimization.
[0093] In one embodiment, Figure 3 As shown, a display effect processing method based on machine learning is provided, and the method is applied to a display device as an example for explanation. The display device includes an environment perception module, a user perception module, a machine learning module and a display adjustment module, including the following steps:
[0094] Step S31, obtaining real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, environmental color temperature and change rate.
[0095] Step S32, capturing and analyzing the viewing behavior of the current user, wherein the viewing behavior includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time.
[0096] Step S33, the current environmental parameters and viewing behavior are processed through a preset deep learning model to obtain a display scheme, which includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation.
[0097] Step S34, controlling the display screen to display with the display scheme to obtain a display effect suitable for the current environment.
[0098] Step S35, receiving the feedback result of the current user on the display effect corresponding to the display scheme. The feedback result includes image quality rating, image quality comparison feeling and picture preference selection, which can facilitate users to quickly and accurately feedback their actual feelings and their own needs.
[0099] Specifically, the step S35 includes:
[0100] Display current display parameters and transition mode to the current user, where the transition mode includes one or more combinations of smooth, fast and intelligent prediction;
[0101] Receive feedback from the current user on the display effect corresponding to the display scheme.
[0102] Step S36, according to the feedback result, the preset deep learning model is optimized and updated. Continuously optimizing the model can improve the display effect generated by the system to better meet the actual needs of customers.
[0103] In the above-mentioned display effect processing method based on machine learning, by obtaining real-time current environmental parameters; capturing and analyzing the current user's viewing behavior; processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme; controlling the display screen to display with the display scheme to obtain a display effect suitable for the current environment; receiving the current user's feedback on the display effect corresponding to the display scheme; and optimizing and updating the preset deep learning model according to the feedback results. The present invention can adjust the display parameters of the display screen in combination with the real-time current environmental parameters and viewing behavior, and continuously optimize the model according to user feedback, that is, it can achieve the purpose of adjusting the optimal display effect according to different environments and users.
[0104] It should be understood that although Figure 1-3 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1-3 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.
[0105] In one embodiment, Figure 4As shown, a display effect processing device based on machine learning is provided, including: an environment perception module 41, a user perception module 42, a machine learning module 43 and a display adjustment module 44, wherein:
[0106] The environment perception module 41 is used to obtain real-time current environment parameters, wherein the current environment parameters include any one or a combination of brightness, environment color temperature and change rate;
[0107] The user perception module 42 is used to capture and analyze the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time;
[0108] A machine learning module 43, configured to process the current environment parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation;
[0109] The display adjustment module 44 is used to control the display screen to display according to the display scheme to obtain a display effect suitable for the current environment.
[0110] Furthermore, the device also includes:
[0111] The model training module is used to train the preset deep learning model.
[0112] Furthermore, the model training module includes:
[0113] Model loading unit, used to load pre-trained models and set user preferences;
[0114] A data collection unit, used to obtain historical environmental parameters and historical viewing behaviors;
[0115] A data screening unit, configured to perform preliminary screening of the historical environmental parameters and historical viewing behaviors through the pre-trained model to remove abnormal data and duplicate data, and to supplement missing data through preset probability calculation;
[0116] A data normalization unit, used for normalizing the filtered historical environment parameters and historical viewing behaviors;
[0117] A feature selection unit, configured to select original features with a high correlation with a target variable from the normalized historical environment parameters and historical viewing behaviors, wherein the target variable corresponds to a parameter in the display scheme;
[0118] A feature encoding unit, used to classify and encode the original features and convert them into numerical features, wherein the classification basis includes user preferences;
[0119] A feature transformation unit, used to transform the original feature, wherein the transformation includes a logarithmic transformation or a square root transformation;
[0120] A feature dimension reduction unit, used to reduce the number of the original features by a dimension reduction method;
[0121] A model training unit is used to train a regression model using the historical environmental parameters and historical viewing behaviors after dimensionality reduction to obtain the preset deep learning model, wherein the regression model includes a neural network model or a support vector machine (Support Vector Machine, SVM).
[0122] Furthermore, the model training module also includes:
[0123] A model evaluation unit is used to evaluate the preset deep learning model using a test data set, wherein the evaluation indicators include one or more combinations of mean square error, root mean square error, and R square.
[0124] Furthermore, the model training module also includes:
[0125] A model optimization unit is used to optimize the preset deep learning model according to the evaluation result, and the optimization includes model parameter optimization and model structure optimization.
[0126] Furthermore, the display adjustment module 44 includes:
[0127] A scheme generating unit, configured to generate a real-time display scheme according to the display scheme and in combination with a corresponding preset threshold, wherein the real-time display scheme includes one or more combinations of display parameters, adjustment speed, and adjustment mode;
[0128] The first adjustment unit is used to dynamically adjust the display parameters of the display screen according to the real-time display solution.
[0129] Furthermore, the display adjustment module 44 further includes:
[0130] An effect monitoring unit, used to monitor the performance of the display screen to adjust the display effect in real time;
[0131] The second adjustment unit is used to adjust the display of the display screen according to a preset solution if the performance is monitored to be poor or the user feedback shows that the current display effect is not good.
[0132] Furthermore, the device also includes:
[0133] The user feedback module is used to receive the feedback result of the current user on the display effect corresponding to the display scheme.
[0134] Furthermore, the user feedback module includes:
[0135] A parameter display unit, used to display current display parameters and transition modes to the current user, wherein the transition modes include one or more combinations of smooth, fast and intelligent prediction;
[0136] The feedback receiving unit is used to receive the feedback result of the current user on the display effect corresponding to the display scheme.
[0137] Furthermore, the device also includes:
[0138] The model iteration module is used to optimize and update the preset deep learning model according to the feedback results.
[0139] For the specific definition of the display effect processing device based on machine learning, please refer to the definition of the display effect processing method based on machine learning above, which will not be repeated here. Each module in the above-mentioned display effect processing device based on machine learning can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the display device in the form of hardware, or can be stored in the memory in the display device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0140] In one embodiment, a display device is provided. The display device may be a server, a computer, an iPad, a tablet, a mobile phone or other display device. The internal structure diagram thereof may be as follows: Figure 5 As shown. The display device includes a processor and a memory connected via a system bus. The processor of the display device is used to provide computing and control capabilities. The memory of the display device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores a computer program. The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, a display effect processing method based on machine learning is implemented.
[0141] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the display device to which the scheme of the present application is applied. The specific display device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0142] In one embodiment, a display device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:
[0143] Acquire real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, ambient color temperature and change rate;
[0144] Capturing and analyzing the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time;
[0145] Processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation;
[0146] The display screen is controlled to display with the display scheme to obtain a display effect suitable for the current environment.
[0147] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:
[0148] Acquire real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, ambient color temperature and change rate;
[0149] Capturing and analyzing the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time;
[0150] Processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation;
[0151] The display screen is controlled to display with the display scheme to obtain a display effect suitable for the current environment.
[0152] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0153] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0154] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A display effect processing method based on machine learning, characterized in that: The method comprises: Acquire real-time current environmental parameters, wherein the current environmental parameters include any one or a combination of brightness, ambient color temperature and change rate; Capturing and analyzing the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time; Processing the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation; The display screen is controlled to display with the display scheme to obtain a display effect suitable for the current environment.
2. The method according to claim 1, characterized in that The method further comprises: Train the preset deep learning model.
3. The method according to claim 2, characterized in that The training preset deep learning model includes: Load pre-trained models and set user preferences; Obtain historical environment parameters and historical viewing behaviors; Preliminarily screening the historical environmental parameters and historical viewing behaviors through the pre-trained model to remove abnormal data and duplicate data, and supplement missing data through preset probability calculation; Normalizing the filtered historical environment parameters and historical viewing behaviors; Selecting original features with a high correlation with a target variable from the normalized historical environmental parameters and historical viewing behaviors, wherein the target variable corresponds to a parameter in the display scheme; Classify and encode the original features and convert them into numerical features, wherein the classification basis includes user preferences; Transforming the original features, wherein the transformation includes logarithmic transformation or square root transformation; Reducing the number of the original features by a dimensionality reduction method; The regression model is trained using the historical environmental parameters and historical viewing behaviors after dimensionality reduction to obtain the preset deep learning model, and the regression model includes a neural network model or a support vector machine (SVM).
4. The method according to claim 3, characterized in that The training preset deep learning model also includes: The preset deep learning model is evaluated using a test data set, wherein the evaluation indicators include one or more combinations of mean square error, root mean square error, and R square.
5. The method according to claim 4, characterized in that The training preset deep learning model also includes: The preset deep learning model is optimized according to the evaluation result, and the optimization includes model parameter optimization and model structure optimization.
6. The method according to claim 1, characterized in that The method of controlling the display screen to display with the display scheme to obtain a display effect adapted to the current environment includes: According to the display scheme, combined with the corresponding preset threshold, a real-time display scheme is generated, wherein the real-time display scheme includes one or more combinations of display parameters, adjustment speed, and adjustment mode; According to the real-time display scheme, the display parameters of the display screen are dynamically adjusted.
7. The method according to claim 6, characterized in that After dynamically adjusting the display parameters of the display screen according to the real-time display scheme, the method includes: Real-time monitoring of the display screen to adjust the performance of the display effect; If the performance is monitored to be poor or the user feedback shows that the current display effect is not good, the display of the display screen is adjusted according to a preset solution.
8. The method according to claim 1, characterized in that After controlling the display screen to display with the display scheme to obtain a display effect adapted to the current environment, the method further includes: Receive feedback from the current user on the display effect corresponding to the display scheme.
9. The method according to claim 8, characterized in that The receiving of the feedback result of the current user on the display effect corresponding to the display scheme includes: Display current display parameters and transition mode to the current user, where the transition mode includes one or more combinations of smooth, fast and intelligent prediction; Receive feedback from the current user on the display effect corresponding to the display scheme.
10. The method according to claim 8, characterized in that The method further comprises: According to the feedback results, the preset deep learning model is optimized and updated.
11. A display effect processing device based on machine learning, characterized in that: The device comprises: An environmental perception module is used to obtain real-time current environmental parameters, which include any one or a combination of brightness, environmental color temperature and change rate; A user perception module is used to capture and analyze the current user's viewing behavior, which includes any one or a combination of sitting posture, distance, viewing angle, eye tracking and viewing time; A machine learning module, configured to process the current environmental parameters and viewing behavior through a preset deep learning model to obtain a display scheme, wherein the display scheme includes any one or a combination of brightness, color balance, color temperature, gamma curve, and natural saturation; The display adjustment module is used to control the display screen to display according to the display scheme to obtain a display effect suitable for the current environment.
12. A display device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.
13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.
