Screen brightness adjusting method and device, electronic equipment, medium and product

Abstract

The embodiment of the invention discloses a screen brightness adjustment method and device, electronic equipment, a storage medium and a product, and the method comprises the steps: obtaining brightness adjustment reference information related to the brightness adjustment of a target screen in the operation process of an application program; performing information feature processing on information items of each information dimension in the brightness adjustment reference information to obtain corresponding information feature vectors; determining a target multi-modal dynamic graph based on an information item of each information dimension in the brightness adjustment reference information; and inputting the information feature vector and the graph structure feature of the target multi-modal dynamic graph into a pre-trained screen brightness adjustment prediction model to obtain a corresponding model prediction result, and adjusting the brightness of the target screen according to the model prediction result. According to the technical scheme provided by the embodiment of the invention, the screen brightness can be adjusted in combination with the multi-dimensional information, so that the personalized screen use requirement of a user is met.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of computer, and particularly, to a screen brightness adjustment method and device, electronic equipment, medium and product. BACKGROUND

[0002] In the process of application of intelligent devices, the brightness of the screen will affect the user experience. The brightness of the display screen can be automatically adjusted according to the ambient brightness, or manually adjusted by the user.

[0003] At present, in the process of automatic adjustment of the brightness of the display screen, only the ambient brightness factor or other single factor is considered, the screen brightness adjustment rule is relatively simple, and multiple dimensions of brightness adjustment influencing factors cannot be comprehensively considered, which cannot meet the personalized needs of users. SUMMARY

[0004] The present disclosure provides a screen brightness adjustment method and device, electronic equipment, medium and product, which can analyze the brightness adjustment reference information based on multiple dimensions to determine the screen brightness adjustment strategy, so that the screen brightness adjustment result can better meet the user's use demand.

[0005] In a first aspect, the embodiments of the present disclosure provide a screen brightness adjustment method, which comprises:

[0006] In the process of running the application program, brightness adjustment reference information associated with the brightness adjustment of the target screen is obtained; wherein the target screen is the display screen of the terminal device running the application program;

[0007] The information items in each information dimension of the brightness adjustment reference information are respectively processed for information features to obtain corresponding information feature vectors;

[0008] A target multi-modal dynamic graph is determined based on the information items in each information dimension of the brightness adjustment reference information; wherein the target multi-modal dynamic graph is used to represent the association relationship between the brightness adjustment reference information;

[0009] The information feature vectors and the graph structure features of the target multi-modal dynamic graph are input into a pre-trained screen brightness adjustment prediction model to obtain corresponding model prediction results, and the brightness of the target screen is adjusted according to the model prediction results.

[0010] In a second aspect, the embodiments of the present disclosure also provide a screen brightness adjustment device, which comprises:

[0011] The data acquisition module is configured to obtain, in the process of running the application program, brightness adjustment reference information associated with the brightness adjustment of the target screen; wherein the target screen is the display screen of the terminal device running the application program;

[0012] The first data processing module is configured to perform information feature processing on each information item in each information dimension of the brightness adjustment reference information, to obtain a corresponding information feature vector.

[0013] The second data processing module is configured to determine a target multi-modal dynamic graph based on the information item in each information dimension of the brightness adjustment reference information, wherein the target multi-modal dynamic graph is used to represent the correlation between the brightness adjustment reference information.

[0014] The data analysis and brightness adjustment module is configured to input the information feature vector and the graph structure feature of the target multi-modal dynamic graph into a pre-trained screen brightness adjustment prediction model, to obtain a corresponding model prediction result, and to adjust the brightness of the target screen according to the model prediction result.

[0015] In a third aspect, the embodiments of the present disclosure further provide an electronic device, which comprises:

[0016] one or more processors;

[0017] a storage device configured to store one or more programs,

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the screen brightness adjustment method according to any of the embodiments of the present disclosure.

[0019] In a fourth aspect, the embodiments of the present disclosure further provide a storage medium containing computer executable instructions, which, when executed by a computer processor, are used to perform the screen brightness adjustment method according to any of the embodiments of the present disclosure.

[0020] In a fifth aspect, the embodiments of the present disclosure further provide a computer program product comprising a computer program, which, when executed by a processor, implements the screen brightness adjustment method according to any of the embodiments of the present disclosure.

[0021] In the embodiments of the present disclosure, during the running of an application, brightness adjustment reference information associated with the brightness adjustment of a target screen is acquired, wherein the target screen is a display screen of a terminal device running the application; information feature processing is respectively performed on information items of each information dimension in the brightness adjustment reference information to obtain corresponding information feature vectors; a target multi-modal dynamic graph is determined based on the information items of each information dimension in the brightness adjustment reference information, wherein the target multi-modal dynamic graph is used to represent the association relationship between the brightness adjustment reference information; the information feature vectors and the graph structure features of the target multi-modal dynamic graph are input into a pre-trained screen brightness adjustment prediction model to obtain corresponding model prediction results, and the brightness of the target screen is adjusted according to the model prediction results. The technical scheme of the embodiments of the present disclosure solves the problem of single condition parameter and lack of personalization in the display screen brightness adjustment in the prior art, realizes screen brightness adjustment with more comprehensive condition parameters, that is, analysis based on multi-dimensional brightness adjustment reference information to determine a screen brightness adjustment strategy, so that the screen brightness adjustment result better meets the use demand of a user. BRIEF DESCRIPTION OF DRAWINGS

[0022] The above and other features, advantages, and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, same or similar reference numerals are used to represent same or similar elements. It should be understood that the drawings are schematic, and the original and elements are not necessarily drawn according to the scale.

[0023] Figure 1 is a screen brightness adjustment method flowchart provided by an embodiment of the present disclosure;

[0024] Figure 2 is a screen brightness adjustment method flowchart provided by an embodiment of the present disclosure;

[0025] Figure 3 is a screen brightness adjustment method flowchart provided by an embodiment of the present disclosure;

[0026] Figure 4 is a screen brightness adjustment method application example diagram provided by an embodiment of the present disclosure;

[0027] Figure 5 is a structure diagram of a screen brightness adjustment device provided by an embodiment of the present disclosure;

[0028] Figure 6 is a structure diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] Embodiments of the present disclosure will be described below in greater detail with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be construed as being limited to the embodiments set forth herein, but rather, the embodiments are provided so that the present disclosure can be more thoroughly and completely understood. It should be understood that the drawings of the present disclosure and the embodiments are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0030] It should be understood that each of the steps recited in the method embodiments of the present disclosure can be performed in different orders and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0031] The term "comprising" and variations thereof as used herein are open-ended, that is, "comprising but not limited to." The term "based on" is "based, at least in part, on." The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments." Related terms are defined in the description that follows.

[0032] It should be noted that the concepts of "first", "second", etc. mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0033] It should be noted that the modification of "one" or "multiple" mentioned in the present disclosure is illustrative and not limiting, and those skilled in the art should understand that, unless otherwise explicitly indicated in the context, it should be understood as "one or more".

[0034] It can be understood that, before using the technical solutions disclosed by the embodiments of the present disclosure, the type, scope of use, use scenario, etc. of the personal information involved in the present disclosure should be informed to the user and the authorization of the user should be obtained in a proper manner according to relevant laws and regulations.

[0035] For example, when responding to the active request of the user, prompt information is sent to the user to explicitly prompt the user that the operation requested to be performed will require the acquisition and use of the personal information of the user. Thus, the user can voluntarily choose whether to provide personal information to the software or hardware such as electronic devices, application programs, servers or storage media that perform the operation of the technical solutions of the present disclosure according to the prompt information.

[0036] As an optional but non-limiting implementation, in response to receiving the active request of the user, the manner of sending the prompt information to the user may be, for example, a pop-up window manner in which the prompt information may be presented in a textual manner. In addition, the pop-up window may also carry a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0037] It can be understood that the above notification and user authorization obtaining process is only illustrative and does not limit the implementation of the present disclosure, and other manners meeting the relevant laws and regulations can also be applied to the implementation of the present disclosure.

[0038] It can be understood that the data involved in the technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of the relevant laws and regulations and the relevant provisions.

[0039] Figure 1 A flowchart of a screen brightness adjustment method provided by the embodiment of the present disclosure, the embodiment of the present disclosure is applicable to the scenario of automatically adjusting the brightness of a display screen during the use of an application program. The method can be executed by a screen brightness adjustment device, which can be implemented in the form of software and / or hardware. Optionally, the screen brightness adjustment device can be implemented by an electronic device, which can be a mobile terminal, a PC terminal, or a server, etc.

[0040] As shown in Figure 1 , the screen brightness adjustment method comprises:

[0041] S110, obtaining brightness adjustment reference information associated with the brightness adjustment of a target screen during the running of an application program; wherein the target screen is a display screen of a terminal device running the application program.

[0042] When an application program relies on a terminal device to run, the terminal device can display the information content provided by the application program through the screen of the terminal device. For example, an application program with a video playing function can play video content through the screen or display any form of information through the screen.

[0043] The brightness of the target screen of the terminal device running the application program usually affects the experience and / or duration of the user browsing information.

[0044] The target screen is the display screen of the terminal device running the application program, and the brightness adjustment reference information comprises multi-dimensional and multi-modal information. Different information dimensions can include multiple information items. For example, multiple information items corresponding to the terminal device dimension, information items corresponding to the environment dimension used by the target screen, information items corresponding to the user's brightness setting or adjustment record of the target screen, information items corresponding to the picture content of the target screen, and information items corresponding to the screen use state dimension, etc.

[0045] The brightness adjustment reference information associated with the brightness adjustment process of the target screen can be acquired in the process of running the application. The brightness adjustment reference information for the current brightness adjustment analysis can be acquired by sliding a time window. The speed of the sliding time window can reflect the time granularity of the screen brightness adjustment analysis, which can be a second-level analysis frequency or a minute-level analysis frequency. The preset length of the sliding time window can be set according to the analysis requirement.

[0046] In S120, the information items in each information dimension of the brightness adjustment reference information are respectively processed to obtain corresponding information feature vectors.

[0047] The information items in each information dimension of the brightness adjustment reference information are respectively processed to enable the neural network model to recognize, understand and analyze the brightness adjustment reference information.

[0048] For different modal brightness adjustment reference information, an information feature processing method matched with the corresponding information modal can be used. For example, for image information, a pixel analysis method can be used for feature extraction, or a neural network model can be used for image feature extraction. For a series of time-related sequence data, a model capable of extracting time dimension features (such as a long short-term memory neural network) can be used. Mathematical statistics, modeling and other mathematical methods can also be used for information feature calculation.

[0049] In S130, a target multi-modal dynamic graph is determined based on the information items in each information dimension of the brightness adjustment reference information. The target multi-modal dynamic graph is used to represent the correlation between the brightness adjustment reference information.

[0050] The target multi-modal dynamic graph is used to represent the correlation between the brightness adjustment reference information. The graph structure feature can reflect the importance of different graph nodes for screen brightness adjustment, which serves as an information reference for the neural network model to analyze the brightness adjustment reference information.

[0051] The target multi-modal dynamic graph includes graph nodes and edges. Each graph node corresponds to an information dimension of the brightness adjustment reference information. According to the division of the information dimension, the target multi-modal dynamic graph contains several information dimensions, and the target multi-modal dynamic graph contains several graph nodes. Further, a graph node can be connected to another graph node based on the correlation between the information items in the information dimension corresponding to the graph nodes.

[0052] For example, the brightness adjustment reference information corresponding to a certain time point, the environmental information indicates that the target screen uses the environment brightness is dark, and the user brightness adjustment operation is to reduce the brightness amplitude by 15%, which indicates that the graph node corresponding to the environmental information dimension and the graph node corresponding to the user operation record dimension. The rule for identifying the association relationship between the graph nodes can be a pre-set logical judgment rule. After the graph nodes and the edges between the graph nodes are determined, the corresponding target multi-modal dynamic graph can be obtained to indicate that in the current screen brightness adjustment analysis, the target multi-modal dynamic graph that meets the current screen state of the user is obtained.

[0053] It should be noted that S120 and S130 do not have strict sequence requirements, and can also be performed simultaneously for brightness adjustment reference information processing. The target and method of data processing between the two are different.

[0054] S140, input the information feature vector and the graph structure feature of the target multi-modal dynamic graph into the pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction result, and adjust the brightness of the target screen according to the model prediction result.

[0055] The graph structure feature of the target multi-modal dynamic graph can be a description feature of the corresponding nodes and edges in the brightness adjustment influencing factor relationship graph, including at least one of the node center degree, the path distance corresponding to the edge, the node weight and the weight of the edge, and the like.

[0056] The information feature vector and the graph structure feature of the target multi-modal dynamic graph are input into the pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction result.

[0057] The screen brightness adjustment prediction model is trained based on a large number of multi-dimensional, multi-modal information samples related to screen brightness adjustment collected in screen use experience or experiments. The learning process of the screen brightness adjustment prediction model is based on the information feature vector of the information sample related to the screen brightness adjustment and the corresponding multi-modal dynamic graph. In this process, multi-dimensional features are involved, and the trained model has strong generalization ability and stability.

[0058] The information feature vector and the graph structure feature in the updated brightness adjustment influencing factor relationship graph are input into the pre-trained application usage time length prediction model, and the model prediction result obtained can be an adjustment amplitude prediction result of the screen brightness adjustment. The brightness of the target screen is directly adjusted according to the model prediction result.

[0059] In the embodiment, the brightness adjustment reference information associated with the brightness adjustment of the target screen is acquired during the running of the application program; the target screen is the display screen of the terminal device running the application program; the information items in each information dimension of the brightness adjustment reference information are respectively subjected to information feature processing to obtain corresponding information feature vectors; the target multi-modal dynamic graph is determined based on the information items in each information dimension of the brightness adjustment reference information; the target multi-modal dynamic graph is used to represent the association relationship between the brightness adjustment reference information; the information feature vectors and the graph structure features of the target multi-modal dynamic graph are input into the pre-trained screen brightness adjustment prediction model to obtain corresponding model prediction results, and the brightness of the target screen is adjusted according to the model prediction results. The technical scheme of the embodiment of the present disclosure solves the problem of single condition parameter of display screen brightness adjustment in the prior art and lacks personalization, realizes screen brightness adjustment with more comprehensive condition parameters, that is, analysis based on multi-dimensional brightness adjustment reference information to determine a screen brightness adjustment strategy, so that the screen brightness adjustment result better meets the use demand of the user.

[0060] Figure 2 A flowchart of a screen brightness adjustment method provided by an embodiment of the present disclosure is shown in FIG. 1. The automatic analysis and adjustment process of the display screen brightness adjustment is further explained and described based on the above embodiment. The method can be executed by a screen brightness adjustment device. The device can be implemented in the form of software and / or hardware, and can be implemented by an electronic device, which can be a mobile terminal, a PC terminal or a server, etc.

[0061] As shown in FIG. 1, the screen brightness adjustment method includes the following steps. Figure 2

[0062] S210, in the process of running the application program, brightness adjustment reference information associated with the brightness adjustment of the target screen is acquired; the target screen is the display screen of the terminal device running the application program.

[0063] In an optional implementation, the brightness adjustment reference information can include brightness adjustment record sequence information for the brightness adjustment of the target screen, device information of the terminal device, environment feature information of the environment where the terminal device is located, picture feature information of the display picture of the target screen and use state information of the target screen.

[0064] ​The brightness adjustment record sequence information can be a regular manual brightness adjustment value, a specific numerical value of screen brightness set by a user through manual operation; an automatic brightness adjustment value based on ambient light, a brightness numerical value automatically adjusted by the device according to the intensity of ambient light; a brightness adjustment value in a specific application scenario, for example, a brightness value adjusted by a user or automatically adjusted by the system in different scenarios such as reading applications, game applications, video playing applications, and the like; a brightness adjustment value in different time periods, such as brightness settings in different time periods such as daytime, nighttime, morning, afternoon, and the like; a brightness adjustment value in different geographic locations, brightness values set in different places (such as indoors, outdoors, different cities, and the like); and a brightness adjustment value in different use scenarios (such as work, entertainment, learning, and the like).

[0065] The device information of the terminal device can be an energy consumption mode, including different device energy consumption settings such as an energy saving mode and a high performance mode; a battery health degree, such as reflecting the remaining life of the battery, the number of charging cycles, the capacity attenuation of the battery, and the like; a screen aging index, measuring the degree of decline of the screen display quality with the use time, for example, the change in brightness uniformity, color deviation, and the like; a screen resolution, such as different resolution settings such as high definition, full high definition, and ultra high definition; a screen refresh rate, commonly having different refresh rates such as 60 Hz, 90 Hz, 120 Hz, and the like; and a screen color rendering capability evaluation, including indicators such as color accuracy, color gamut coverage (such as sRGB, Adobe RGB, and the like), and contrast.

[0066] The environmental feature information of the environment in which the terminal device is located can be an ambient light level, such as measuring the light intensity of the surrounding environment, including different degrees such as strong light, weak light, and medium light.

[0067] The picture feature information of the display picture of the target screen can include a video picture average brightness value, a dark area proportion, a color vividness, a video frame rate, a picture definition, and a video encoding format influence coefficient. Among them, the video picture average brightness value represents the average brightness level of the entire video picture; the dark area proportion represents the proportion of the darker part in the picture; the color vividness represents the richness and brightness of the color of the video picture; the video frame rate represents the number of video frames displayed per second, such as 24 frames, 30 frames, 60 frames, and the like; the picture definition includes the sharpness and detail performance of the image; the video encoding format influence coefficient includes the influence degree of different video encoding formats (such as H.264, H.265, and the like) on the picture quality; and the picture dynamic range is, for example, high dynamic range (HDR) and standard dynamic range (SDR).

[0068] The use state information of the target screen includes the frequency and intensity of screen operation in a specific time period, the duration of using the eye protection mode, the duration of continuous use of the same application, the habit preference for screen brightness adjustment, user behavior posture, and the like. Among them, the frequency and intensity of screen operation in a specific time period, for example, can be the number of times of clicking the screen and the intensity of clicking in an hour. The duration of using the eye protection mode is the duration of turning on the eye protection mode. The duration of continuous use of the same application indicates the length of time of continuously using a certain specific application (such as a social media application, a game application, and the like). An exemplary habit preference for screen brightness adjustment can be a tendency to always prefer a higher or lower brightness. The user behavior posture can be the posture of the handheld device (landscape, portrait), the distance of watching the screen, and the like, and whether the device is used in movement (such as walking, taking a vehicle, and the like).

[0069] S220, information feature processing is performed on each information item in each information dimension in the brightness adjustment reference information respectively to obtain a corresponding information feature vector.

[0070] For the information item content of each information dimension, information feature processing can be performed in the following manner.

[0071] For information feature processing of the picture feature information of the display picture of the target screen, image decomposition can be performed on the display picture to obtain a plurality of different sub-band images of preset dimensions and frequencies, and the sub-band images are input into a preset convolutional neural network for image feature extraction to obtain an information feature vector corresponding to the picture feature information of the display picture of the target screen. Based on a bidirectional long short-term memory network, the brightness adjustment record sequence information and the use state information can be encoded respectively to obtain information feature vectors corresponding to the brightness adjustment record sequence information and the use state information. Based on an autoencoder, the device information and the environmental feature information can be encoded respectively to obtain information feature vectors corresponding to the device information and the environmental feature information.

[0072] S230, according to the association relationship between the information items in each information dimension in the brightness adjustment reference information, the node degree of the graph node of each information dimension pair, the clustering coefficient and the path length between the nodes are determined.

[0073] Each information dimension corresponds to a graph node, and edges between the corresponding graph nodes can be established according to the association relationship between each information item in the brightness adjustment reference information, and then the node degree corresponding to the node and the clustering coefficient and the path length between the nodes can be determined.

[0074] The node degree refers to the number of edges connected to a node. It reflects the degree of connection of the node in the network. Nodes with high node degree usually have a more important position in the network because they have direct contact with more nodes.

[0075] The clustering coefficient between nodes measures the degree of aggregation of nodes in the network. It represents the ratio of the actual number of edges between the neighbors of a node to the maximum number of edges that can exist. A high clustering coefficient means that the neighbors of a node have a high density of connections, and the network exhibits a clustered structure.

[0076] The path length refers to the number of edges on the shortest path between two nodes. The average path length is the average of the path lengths between all pairs of nodes in the network. A shorter path length means that information can propagate faster in the network.

[0077] S240, according to the data relationship between the node degree, clustering coefficient and path length and the reference node degree, reference clustering coefficient and reference path length in the preset reference multimodal dynamic graph, determine the node weight of the corresponding graph node and the edge weight corresponding to the connection line between the graph nodes, and obtain the target multimodal dynamic graph.

[0078] Among them, the preset reference multimodal dynamic graph can be a reference information obtained by analyzing sample data based on the training screen brightness adjustment prediction model, reflecting the relationship and importance between the factors in the screen brightness adjustment process. Each graph node in the preset reference multimodal dynamic graph has a corresponding reference node weight calculated and analyzed and assigned, and the edges between nodes have reference edge weights.

[0079] In the application process of the screen brightness adjustment method, personalized brightness adjustment reference information can be obtained according to the current screen use state of the current user, and then the target multimodal dynamic graph corresponding to the current screen use state of the current user is updated, which instantly adapts to environmental or scene changes and more accurately analyzes the brightness adjustment reference information for screen brightness modulation.

[0080] According to the data relationship between the node degree, clustering coefficient and path length and the reference node degree, reference clustering coefficient and reference path length in the preset reference multimodal dynamic graph, the graph node weight of the corresponding graph node and / or the edge weight corresponding to the connection line between the nodes can be increased or decreased, that is, the target multimodal dynamic graph can be determined. For example, if the node degree of a certain graph node increases compared to the reference node degree, the node weight of the corresponding node can be increased accordingly.

[0081] S250, input the information feature vector and the graph structure feature of the target multimodal dynamic graph into the screen brightness adjustment prediction model pre-trained, and obtain the corresponding model prediction result.

[0082] The information feature vector and the graph structure feature in the updated brightness adjustment influence factor relationship graph are input into the application program use time length prediction model or the screen brightness adjustment prediction model pre-trained, and the corresponding model prediction result is obtained.

[0083] S260, when the model prediction result is the use time length prediction result of the application program, determining a target screen brightness corresponding to the model prediction result according to an application use time length and screen brightness mapping relationship, and adjusting the brightness of the target screen to the target screen brightness; and when the model prediction result is the screen brightness adjustment amplitude prediction result, increasing or decreasing the brightness of the target screen by the brightness adjustment amplitude corresponding to the model prediction result.

[0084] The application use time length and screen brightness mapping relationship can be determined based on statistical values, and the corresponding relationship between the application use time length and the screen brightness value.

[0085] The technical solution of the embodiments of the present disclosure obtains the brightness adjustment reference information associated with the brightness adjustment of the target screen in the process of running the application program, wherein the target screen is the display screen of the terminal device running the application program; respectively processes the information items in each information dimension in the brightness adjustment reference information to obtain the corresponding information feature vectors; determines the node degree of the graph node, the clustering coefficient and the path length between nodes of each information dimension pair according to the association relationship between the information items in each information dimension in the brightness adjustment reference information; determines the node weight of the corresponding graph node and the edge weight corresponding to the connection line between graph nodes according to the data relationship between the node degree, the clustering coefficient and the path length and the reference node degree, the reference clustering coefficient and the reference path length in the preset reference multi-modal dynamic graph, to obtain a target multi-modal dynamic graph; inputs the information feature vectors and the graph structure features of the target multi-modal dynamic graph into the screen brightness adjustment prediction model pre-trained to obtain the corresponding model prediction result; when the model prediction result is the use time length prediction result of the application program, determines the target screen brightness corresponding to the model prediction result according to the application use time length and screen brightness mapping relationship, and adjusts the brightness of the target screen to the target screen brightness; and when the model prediction result is the screen brightness adjustment amplitude prediction result, increases or decreases the brightness of the target screen by the brightness adjustment amplitude corresponding to the model prediction result. The technical solution of the embodiments of the present disclosure solves the problem of single condition parameter of display screen brightness adjustment in the prior art and lacks of personalization, realizes screen brightness adjustment with more comprehensive condition parameters, i.e., analyzes the brightness adjustment reference information in multiple dimensions to determine the screen brightness adjustment strategy, so that the screen brightness adjustment result better meets the use requirements of users.

[0086] Figure 3A flowchart of a screen brightness adjustment method provided by an embodiment of the present disclosure is shown in the figure. Based on the above embodiment, the process of training the application usage duration prediction model or the screen brightness adjustment prediction model is further explained. The neural network model can have improved prediction ability, generalization, adaptability, and stability. The method can be performed by a screen brightness adjustment device, which can be implemented in software and / or hardware. Optionally, the device can be implemented by an electronic device, such as a mobile terminal, a PC terminal, or a server.

[0087] As shown in Figure 3 The screen brightness adjustment method includes:

[0088] In S310, brightness adjustment reference information sample data is obtained, and a graph node is set for each information dimension in the brightness adjustment reference information sample data. The connection line between the corresponding graph nodes is determined based on the data correlation between each information dimension, and an initial multi-modal dynamic graph is obtained.

[0089] In S310, brightness adjustment reference information sample data is obtained, and a graph node is set for each information dimension in the brightness adjustment reference information sample data. The connection line between the corresponding graph nodes is determined based on the data correlation between each information dimension, and an initial multi-modal dynamic graph is obtained.

[0090] The brightness adjustment reference information sample data includes data of different dimensions that affect screen brightness adjustment, such as terminal device identification information running a specified application, environment information when running the specified application, and user screen brightness adjustment information. The brightness adjustment reference information sample data can be sample data collected under a preset application usage condition, such as various brightness adjustment reference information sample data collected when the application is running for a specified duration. Alternatively, the information includes the amplitude of the user's screen brightness adjustment under different application running conditions.

[0091] According to the brightness adjustment reference information sample data, an initial multi-modal dynamic graph can be constructed to represent and process the complex relationship between multi-modal data, and to represent the relationship between the brightness adjustment influencing factors. The nodes in the initial multi-modal dynamic graph can be determined according to the dimensions of the brightness adjustment reference information sample data. For example, a device feature node represents device-related information, such as screen type, battery status, etc. An environmental data node includes environmental factors such as light intensity and temperature. A user behavior node reflects user interaction with the device, such as screen brightness adjustment and application usage frequency. The connection between nodes can be established based on the correlation of the data. For example, there can be a direct correlation between environmental light intensity and user brightness adjustment behavior. According to the nodes and the connection relationship between the nodes, an initial multi-modal dynamic graph can be obtained.

[0092] The initial multi-modal dynamic graph can represent the relationship state between different graph nodes based on sample data statistical analysis. The graph node weights can be dynamically assigned based on the data contribution and influence of each graph node in the initial multi-modal dynamic graph. For example, if a certain environment node frequently affects the user behavior node, the number of edges between the two nodes is relatively large, the node weight will be increased, and the edge weight will also be increased to emphasize its importance in subsequent model learning.

[0093] S320, denoising the connection relationship between the graph nodes in the initial multi-modal dynamic graph to obtain a reference multi-modal dynamic graph.

[0094] Denoising the connection relationship between the graph nodes in the initial multi-modal dynamic graph to obtain a reference multi-modal dynamic graph.

[0095] In order to improve the accuracy of data processing, so as to improve the robustness of the model obtained by training in complex models. Specifically, in the process of denoising, two denoising processes can be performed in sequence to optimize and clean the connection relationship between nodes in the dynamic graph.

[0096] First, preliminary denoising can be performed, for example, calculating at least one statistical parameter of the connection relationship between the graph nodes in the initial multi-modal dynamic graph, removing abnormal connection relationships indicated by the statistical parameter, and taking the initial multi-modal dynamic graph with the abnormal connection relationships removed as the reference multi-modal dynamic graph. Then, deep denoising can be performed, and the initial multi-modal dynamic graph with the abnormal connections removed can be input into a graph neural network based on an attention mechanism to obtain connection edge weights corresponding to the connection relationship between the graph nodes. Furthermore, the multi-modal dynamic graph with the abnormal connections removed is taken as an updated reference multi-modal dynamic graph, and the reference node weights and the reference edge weights of the connection lines between the graph nodes in the reference multi-modal dynamic graph.

[0097] In the preliminary denoising process, the goal is to identify and remove noisy connections that may be caused by measurement errors, data transmission errors, or external interference. Statistical parameters such as mean, standard deviation, and proportion of outliers can be calculated for each connection. Threshold methods such as Z-score or IQR method are used to identify and remove statistically significant abnormal connections.

[0098] In the deep denoising process, the goal is to adjust and optimize the weights of the remaining connections. Specifically, an attention mechanism can be introduced in the graph neural network to dynamically adjust the weights of the connections between nodes. The attention weight is determined by the importance and contribution of the connection, and can be normalized by a softmax function:

[0099] Here, `score(i,j)` is a scoring function that can be calculated using a small neural network based on connectivity features. The optimization objective is to minimize the connectivity error after denoising, while maintaining the integrity and representativeness of the data. This can be achieved using a loss function that includes a connectivity error term and a regularization term. Among them, y ij It is the actual connection strength. θ represents the connection strength predicted by the model, θ represents the model parameters, and λ represents the regularization parameter.

[0100] S330. Input the information feature vector corresponding to the brightness adjustment reference information sample data and the graph structure features of the benchmark multimodal animation into the model to be trained to obtain the model output results.

[0101] The information feature vector corresponding to the brightness adjustment reference information sample data and the graph structure feature corresponding to the updated baseline brightness adjustment influencing factor relationship graph are input into the model to be trained to obtain the model output result.

[0102] The graph structure features can be at least one of the following: node weights in the updated baseline brightness adjustment influencing factor relationship graph, weights of connecting edges between nodes, node centrality, node clustering coefficient, and average path length between nodes.

[0103] The model to be trained is a neural network model based on a multi-head attention mechanism, which includes attention mechanisms for edge weight information corresponding to the connection lines between graph nodes in the benchmark multimodal animation graph, such as the Edge-InformedAttention strategy.

[0104] The Edge-Informed Attention strategy incorporates information from the edges between nodes. These edges represent the importance of the nodes themselves, as well as the strength of the relationships and interactions between them. In this way, the model being trained not only learns the importance of the features themselves but also understands the importance of the relationships between different nodes, thus better capturing and representing complex data structures.

[0105] Traditionally, attention scores are calculated using transformations of three vectors: query, key, and value. Edge-Informed Attention, however, multiplies the similarity between the query and key by the weights of the corresponding edges.

[0106] Specifically, if the edge weight between node i and node j is ω ij Then their attention scores Attention(Q) i ,K j ) will be adjusted to ωij Attention(Q i ,K j ).

[0107] In some cases, direct connections between nodes can not be sufficient to express the full relationship between them, and the path between nodes can also contain important information. The path-dependent graph attention mechanism considers the indirect relationship between nodes connected by different paths. First, the key paths between nodes in the graph can be identified (such as the shortest path, the strongest path, etc.), and path discovery is performed. Then, path weight calculation is performed, for example, the weight of each path is calculated based on factors such as the weight and length of the edges on the path. Finally, attention score adjustment is performed, and the influence of indirect connection through direct edges and key paths is considered when calculating the attention score between nodes.

[0108] S340, based on the model output result, the sample label data corresponding to the brightness adjustment reference information sample data, and the graph structure feature, the model parameter is updated to obtain a screen brightness adjustment prediction model.

[0109] Based on the model output result, the sample label data corresponding to the brightness adjustment reference information sample data, and the graph structure feature, the model parameter is updated to obtain an application usage time length prediction model and / or a screen brightness adjustment prediction model.

[0110] Wherein, the sample label data is the application usage time length data and / or the brightness adjustment amplitude data corresponding to the brightness adjustment reference information sample data.

[0111] Wherein, the sample label data is the application usage time length data and / or the actual brightness adjustment amplitude data corresponding to the brightness adjustment reference information sample data. The output result of the model trained based on different sample labels is also different. The application usage time length data is an indirect user satisfaction feedback, which can be used to infer the satisfaction degree of brightness setting through application usage time length or adjustment frequency information, as a label. The actual brightness adjustment amplitude data is a direct user feedback in the process of using the application, which is recorded as a label in the sample data collection process.

[0112] Specifically, based on the model output result and the sample label data corresponding to the brightness adjustment reference information sample data, and the graph structure feature, the model parameter updating process can calculate the model learning loss according to the model output result and the sample label data corresponding to the brightness adjustment reference information sample data; determine the regularization loss according to at least one feature data in the graph structure feature; and update the parameters of the model to be trained according to the model learning loss and the regularization loss.

[0113] The regularization loss utilizes the topological structure information of a dynamic graph to constrain the learning process. This regularization method helps the model optimize the accuracy of feature representation and fusion effect while maintaining the data structure attributes. The core idea of ​​structural regularization is to use the graph's structural information as additional guiding information to guide the model's learning process. This method helps the model better understand and utilize the relationships between nodes, thereby improving the accuracy of feature representation and the effect of data fusion. This regularization term is a function based on structural features; for example, the inverse of the node degree centrality can be used as the regularization strength (using the inverse of the node degree centrality as the regularization strength is actually to reduce the model's over-reliance on highly central nodes. In many cases, highly central nodes may receive too much attention in the model, thus masking the contributions of other nodes). During model training, optimization algorithms such as gradient descent are combined with the structural regularization term to continuously adjust the model parameters to minimize the total loss. Structural regularization helps the model being trained to focus not only on the representation of the data but also on the data's internal structure during the learning process, thereby achieving better generalization results. By introducing graph structural information, the model can consider the inherent connections and dependencies of the data during training, thereby improving generalization ability. Structural regularization can reduce the risk of model overfitting, enabling the model to maintain stable performance across different datasets. It makes the model's decision-making process more transparent because the model's output is closely related not only to data features but also to the structural characteristics of the data.

[0114] S350. During the operation of the application, obtain brightness adjustment reference information associated with the brightness adjustment of the target screen; wherein, the target screen is the display screen of the terminal device running the application.

[0115] During application execution, obtain brightness adjustment reference information associated with the target screen's brightness adjustment process.

[0116] The target screen is the display screen of the terminal device running the application. The brightness adjustment reference information includes brightness adjustment record sequence information for adjusting the brightness of the target screen, device information of the terminal device, environmental characteristic information of the environment in which the terminal device is located, image characteristic information of the display screen of the target screen, and usage status information of the target screen.

[0117] S360. Perform information feature processing on each information item in the brightness adjustment reference information to obtain the corresponding information feature vector.

[0118] S370. Determine the target multimodal animation based on the information items of each information dimension in the brightness adjustment reference information; wherein, the target multimodal animation is used to characterize the correlation between the brightness adjustment reference information.

[0119] The application environment in which the trained application uses the duration prediction model and / or screen brightness adjustment prediction model is a dynamic environment. The node weights and the importance of edges between nodes in the baseline multimodal animation graph may change depending on the user or with changes in time and environment, resulting in the target multimodal animation graph. In this embodiment, adaptive edge updates can be performed based on dynamically acquired brightness adjustment reference information.

[0120] The information feature vector and the graph structure features in the updated brightness adjustment influencing factor relationship graph are input into the pre-trained application usage duration prediction model or screen brightness adjustment prediction model to obtain the corresponding model prediction results.

[0121] S380. Input the information feature vector and the graph structure features of the target multimodal animation into the pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction results, and adjust the brightness of the target screen according to the model prediction results.

[0122] The technical solution of this disclosure involves acquiring brightness adjustment reference information sample data, setting graph nodes for each information dimension in the brightness adjustment reference information sample data, and determining the connection lines between corresponding graph nodes based on the data association relationship between each information dimension to obtain an initial multimodal animation; denoising the connection relationships between graph nodes in the initial multimodal animation to obtain a baseline multimodal animation; inputting the information feature vector corresponding to the brightness adjustment reference information sample data and the graph structure features of the baseline multimodal animation into the model to be trained to obtain the model output result; and updating the model parameters based on the model output result, the sample label data corresponding to the brightness adjustment reference information sample data, and the graph structure features to obtain the screen brightness adjustment pre-processing result. The method involves several steps: First, during application execution, acquiring brightness adjustment reference information associated with the brightness adjustment of the target screen. The target screen is the display screen of the terminal device running the application. Information feature processing is performed on each information item in the brightness adjustment reference information to obtain a corresponding information feature vector. A target multimodal animation is determined based on the information items in each information dimension of the brightness adjustment reference information, representing the correlation between brightness adjustment reference information items. The information feature vector and the graph structure features of the target multimodal animation are input into a pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction result. The brightness of the target screen is then adjusted according to the model prediction result. This embodiment of the invention solves the problem of how to comprehensively analyze multi-dimensional brightness adjustment parameters. A neural network model with multimodal input is trained based on sample data to analyze the collected brightness adjustment reference information and adjust the screen brightness according to the model output result. This allows for automatic adjustment of the target screen brightness during application use to meet user needs.

[0123] In a specific example of a screen brightness adjustment method, the implementation process can be found in [reference needed]. Figure 4 The flowchart shown is as follows. First, basic feature data is collected, namely, the user screen brightness adjustment history dataset L (corresponding to brightness adjustment record sequence information), device information vector D (corresponding to device information), environmental data vector N (corresponding to environmental feature information of the environment where the terminal device is located), video image signal feature F (corresponding to image feature information of the display image of the target screen), and user behavior feature state sequence S (corresponding to the usage state information of the target screen).

[0124] Each basic feature data point undergoes data feature processing via autoencoder, wavelet transform, and convolutional neural network, or bidirectional long short-term memory network, to obtain information features. These features are then input into a pre-trained Transformer model based on a multi-head self-attention mechanism. Simultaneously, the basic feature data is also used to construct a dynamic graph (a graph showing the relationship between factors influencing brightness adjustment), and the graph structure features are also input into the Transformer model based on the multi-head self-attention mechanism. The Transformer model based on the multi-head self-attention mechanism also includes Edge-Informed Attention. By combining edge information from the dynamic graph, the Edge-Informed Attention strategy assigns weights to the interactions between different nodes, reflecting their importance and influence in practical applications.

[0125] Each attention mechanism is followed by a feedforward network to further process the features and perform nonlinear transformations on them, enhancing the model's expressive power. This leads to the model's output.

[0126] The Transformer model in this example, based on a multi-head self-attention mechanism, utilizes a dynamic graph to integrate multimodal data from device, environment, and user behavior. This structure not only captures the direct relationships between data but also dynamically adjusts these relationships to adapt to environmental changes and evolving user habits. This allows for personalized screen brightness adjustment services for users.

[0127] Figure 5 This disclosure provides a screen brightness adjustment device suitable for scenarios where screen brightness is adjusted during application use. The screen brightness adjustment device can be implemented in software and / or hardware and can be configured in an electronic device, such as a mobile terminal, a PC, or a server.

[0128] like Figure 5As shown, the screen brightness adjustment device includes: a data acquisition module 410, a first data processing module 420, a second data processing module 430, and a data analysis and brightness adjustment module 440.

[0129] The data acquisition module 410 is used to acquire brightness adjustment reference information associated with the brightness adjustment of the target screen during the operation of the application; wherein the target screen is the display screen of the terminal device running the application; the first data processing module 420 is used to perform information feature processing on the information items of each information dimension in the brightness adjustment reference information to obtain the corresponding information feature vector; the second data processing module 430 is used to determine the target multimodal animation based on the information items of each information dimension in the brightness adjustment reference information; wherein the target multimodal animation is used to represent the correlation between the brightness adjustment reference information; the data analysis and brightness adjustment module 440 is used to input the information feature vector and the graph structure features of the target multimodal animation into a pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction result, and adjust the brightness of the target screen according to the model prediction result.

[0130] The technical solution of this disclosure involves acquiring brightness adjustment reference information associated with the brightness adjustment of a target screen during the operation of an application; wherein the target screen is the display screen of the terminal device running the application; performing information feature processing on each information item of each information dimension in the brightness adjustment reference information to obtain a corresponding information feature vector; determining a target multimodal animation based on the information items of each information dimension in the brightness adjustment reference information; wherein the target multimodal animation is used to characterize the correlation between the brightness adjustment reference information; inputting the information feature vector and the graph structure features of the target multimodal animation into a pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction result, and adjusting the brightness of the target screen according to the model prediction result. The technical solution of this disclosure solves the problem of single and lack of personalization in the conditional parameters of display screen brightness adjustment in the prior art, realizing screen brightness adjustment with more comprehensive conditional parameters, that is, analyzing multi-dimensional brightness adjustment reference information to determine the screen brightness adjustment strategy, making the screen brightness adjustment result more meet the user's needs.

[0131] In one alternative implementation, the second data processing module 430 is specifically used for:

[0132] Based on the correlation between information items in each information dimension of the brightness adjustment reference information, determine the node degree, clustering coefficient between nodes, and path length of the graph nodes for each information dimension pair;

[0133] Based on the data relationship between node degree, clustering coefficient, and path length and the baseline node degree, baseline clustering coefficient, and baseline path length in the preset baseline multimodal animation graph, the node weights of the corresponding graph nodes and the edge weights of the connecting lines between graph nodes are determined to obtain the target multimodal animation graph.

[0134] In one optional implementation, the first data processing module 420 is specifically used for:

[0135] The displayed image in the brightness adjustment reference information is decomposed to obtain sub-band images, and these sub-band images are input into a preset convolutional neural network for image feature extraction to obtain the display image information feature vector of the target screen; and / or,

[0136] Based on a bidirectional long short-term memory network, the brightness adjustment record sequence information and usage status information in the brightness adjustment reference information are encoded separately to obtain the information feature vectors corresponding to the brightness adjustment record sequence information and usage status information; and / or,

[0137] The device information and environmental feature information in the brightness adjustment reference information are encoded by an automatic encoder to obtain the information feature vectors corresponding to the device information and environmental feature information.

[0138] In one alternative implementation, the data analysis and brightness adjustment module 440 is specifically used for:

[0139] When the model prediction result is the application usage duration prediction result, the target screen brightness corresponding to the model prediction result is determined based on the mapping relationship between application usage duration and screen brightness, and the target screen brightness is adjusted to the target screen brightness; and,

[0140] When the model prediction result is the screen brightness adjustment range prediction result, the brightness of the target screen will be increased or decreased according to the brightness adjustment range corresponding to the model prediction result.

[0141] In one optional embodiment, the screen brightness adjustment device further includes a model training module for training a screen brightness adjustment prediction model. The training process of the screen brightness adjustment prediction model includes:

[0142] Obtain sample data of brightness adjustment reference information, set graph nodes for each information dimension in the sample data of brightness adjustment reference information, and determine the connection lines between the corresponding graph nodes based on the data relationship between each information dimension to obtain the initial multimodal animation.

[0143] The connection relationships between nodes in the initial multimodal animation are denoised to obtain the baseline multimodal animation.

[0144] The information feature vector corresponding to the brightness adjustment reference information sample data and the graph structure features of the benchmark multimodal animation are input into the model to be trained to obtain the model output results.

[0145] Based on the model output, the sample label data corresponding to the brightness adjustment reference information sample data, and the graph structure features, the model parameters are updated to obtain the screen brightness adjustment prediction model.

[0146] Among them, the sample label data consists of application usage time data and / or brightness adjustment range data corresponding to the brightness adjustment reference information sample data.

[0147] In one alternative implementation, the model training module can also be used for:

[0148] Calculate at least one statistical parameter of the connection relationship between nodes in the initial multimodal animation graph, remove the abnormal connection relationships indicated by the statistical parameter, and use the initial multimodal animation graph with abnormal connection relationships removed as the baseline multimodal animation graph;

[0149] The baseline multimodal animation is input into an attention-based graph neural network to obtain the baseline node weights of the graph nodes in the baseline multimodal animation and the baseline edge weights of the connecting lines between the graph nodes.

[0150] In one optional implementation, the model to be trained is a neural network model based on a multi-head attention mechanism, and the multi-head attention mechanism includes an attention mechanism for the edge weight information corresponding to the connection lines between graph nodes of the baseline multimodal animation graph.

[0151] In one alternative implementation, the model training module is specifically used for:

[0152] Calculate the model learning loss based on the model output and the sample label data corresponding to the brightness adjustment reference information sample data.

[0153] The regularization loss is determined based on at least one feature data from the graph structure features;

[0154] Based on the model learning loss and regularization loss, the parameters of the model to be trained are updated using gradients.

[0155] The screen brightness adjustment device provided in this disclosure can execute the screen brightness adjustment method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects of executing the method.

[0156] It is worth noting that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of this disclosure.

[0157] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Reference is made below. Figure 6 It illustrates an electronic device suitable for implementing embodiments of the present disclosure (e.g., Figure 6 The diagram below shows the structure of the terminal device or server 500. The terminal device in this embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and vehicle terminals (e.g., vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0158] like Figure 6 As shown, electronic device 500 may include a processing unit (e.g., central processing unit, graphics processor, etc.) 501, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from storage device 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of electronic device 500. The processing unit 501, ROM 502, and RAM 503 are interconnected via bus 504. An edit / output (I / O) interface 505 is also connected to bus 504.

[0159] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 508 including, for example, magnetic tapes, hard disks, etc.; and communication devices 509. Communication device 509 allows electronic device 500 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 An electronic device 500 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0160] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by the processing device 501, it performs the functions defined in the methods of embodiments of this disclosure.

[0161] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0162] The electronic device provided in this embodiment and the screen brightness adjustment method provided in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0163] This disclosure also provides a computer storage medium storing a computer program that, when executed by a processor, implements the screen brightness adjustment method provided in the above embodiments.

[0164] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0165] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0166] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0167] The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to:

[0168] During application execution, obtain brightness adjustment reference information associated with the brightness adjustment of the target screen; where the target screen is the display screen of the terminal device running the application.

[0169] The information items of each information dimension in the brightness adjustment reference information are processed to obtain the corresponding information feature vectors.

[0170] The target multimodal animation is determined based on the information items of each information dimension in the brightness adjustment reference information; wherein, the target multimodal animation is used to characterize the correlation between the brightness adjustment reference information.

[0171] The information feature vector and the graph structure features of the target multimodal animation are input into a pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction results, and the brightness of the target screen is adjusted according to the model prediction results.

[0172] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0173] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0174] The units described in the embodiments of this disclosure can be implemented in software or in hardware. The name of a unit does not necessarily limit the unit itself; for example, the first acquisition unit can also be described as "a unit that acquires at least two Internet Protocol addresses".

[0175] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0176] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0177] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the screen brightness adjustment method provided in any embodiment of this disclosure.

[0178] In implementing a computer program product, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0179] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0180] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0181] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for adjusting screen brightness, characterized in that, include: During the operation of the application, brightness adjustment reference information associated with the brightness adjustment of the target screen is obtained; wherein, the target screen is the display screen of the terminal device running the application; Information feature processing is performed on each information item of each information dimension in the brightness adjustment reference information to obtain the corresponding information feature vector; A target multimodal animation is determined based on the information items of each information dimension in the brightness adjustment reference information; wherein, the target multimodal animation is used to characterize the correlation between the brightness adjustment reference information. The information feature vector and the graph structure features of the target multimodal animation are input into a pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction results, and the brightness of the target screen is adjusted according to the model prediction results.

2. The method according to claim 1, characterized in that, The step of determining the target multimodal animation based on information items of each information dimension in the brightness adjustment reference information includes: Based on the correlation between information items in each information dimension of the brightness adjustment reference information, determine the node degree, clustering coefficient between nodes, and path length of the graph nodes for each information dimension pair; Based on the data relationship between the node degree, the clustering coefficient, and the path length and the benchmark node degree, benchmark clustering coefficient, and benchmark path length in the preset benchmark multimodal animation graph, the node weight of the corresponding graph node and the edge weight of the connecting line between graph nodes are determined to obtain the target multimodal animation graph.

3. The method according to claim 1, characterized in that, The information feature processing is performed on each information item in each information dimension of the brightness adjustment reference information to obtain the corresponding information feature vector, including: The display image in the brightness adjustment reference information is decomposed to obtain a sub-band image, and the sub-band image is input into a preset convolutional neural network for image feature extraction to obtain the display image information feature vector of the target screen; and / or, Based on a bidirectional long short-term memory network, the brightness adjustment record sequence information and usage status information in the brightness adjustment reference information are encoded respectively to obtain information feature vectors corresponding to the brightness adjustment record sequence information and the usage status information; and / or, The device information and environmental feature information in the brightness adjustment reference information are encoded by an autoencoder to obtain the information feature vectors corresponding to the device information and the environmental feature information.

4. The method according to claim 1, characterized in that, The step of adjusting the brightness of the target screen based on the model prediction result includes: When the model prediction result is the application usage duration prediction result, the target screen brightness corresponding to the model prediction result is determined according to the mapping relationship between application usage duration and screen brightness, and the target screen brightness is adjusted to the target screen brightness; and, When the model prediction result is a prediction result of the screen brightness adjustment range, the brightness of the target screen is increased or decreased by the brightness adjustment range corresponding to the model prediction result.

5. The method according to any one of claims 1-4, characterized in that, The training process of the screen brightness adjustment prediction model includes: Obtain sample data of brightness adjustment reference information, set graph nodes for each information dimension in the sample data of brightness adjustment reference information, and determine the connection lines between the corresponding graph nodes based on the data correlation between each information dimension to obtain an initial multimodal animation. The connection relationships between nodes in the initial multimodal animation are denoised to obtain a baseline multimodal animation. The information feature vector corresponding to the brightness adjustment reference information sample data and the graph structure features of the benchmark multimodal animation are input into the model to be trained to obtain the model output result. Based on the model output, the sample label data corresponding to the brightness adjustment reference information sample data, and the graph structure features, the model parameters are updated to obtain the screen brightness adjustment prediction model. The sample label data consists of application usage duration data and / or brightness adjustment range data corresponding to the brightness adjustment reference information sample data.

6. The method according to claim 5, characterized in that, The step of denoising the connection relationships between nodes in the initial multimodal animation to obtain a baseline multimodal animation includes: Calculate at least one statistical parameter of the connection relationship between nodes in the initial multimodal animation graph, remove the abnormal connection relationships indicated by the statistical parameter, and use the initial multimodal animation graph with the abnormal connection relationships removed as the baseline multimodal animation graph; The baseline multimodal animation is input into an attention-based graph neural network to obtain the baseline node weights of the graph nodes and the baseline edge weights of the connecting lines between the graph nodes in the baseline multimodal animation.

7. The method according to claim 5, characterized in that, The model to be trained is a neural network model based on a multi-head attention mechanism, and the multi-head attention mechanism includes an attention mechanism for the edge weight information corresponding to the connection lines between graph nodes of the benchmark multimodal animation graph.

8. The method according to claim 5, characterized in that, Based on the model output, the sample label data corresponding to the brightness adjustment reference information sample data, and the graph structure features, the model parameters are updated, including: The model learning loss is calculated based on the model output and the sample label data corresponding to the brightness adjustment reference information sample data. The regularization loss is determined based on at least one feature data from the graph structure features; The parameters of the model to be trained are updated using gradients based on the model learning loss and the regularization loss.

9. A screen brightness adjustment device, characterized in that, include: The data acquisition module is used to acquire brightness adjustment reference information associated with the brightness adjustment of the target screen during the operation of the application; wherein, the target screen is the display screen of the terminal device running the application; The first data processing module is used to perform information feature processing on each information item of each information dimension in the brightness adjustment reference information to obtain the corresponding information feature vector. The second data processing module is used to determine a target multimodal animation based on the information items of each information dimension in the brightness adjustment reference information; wherein, the target multimodal animation is used to characterize the correlation between the brightness adjustment reference information. The data analysis and brightness adjustment module is used to input the information feature vector and the graph structure features of the target multimodal animation into a pre-trained screen brightness adjustment prediction model to obtain the corresponding model prediction results, and adjust the brightness of the target screen according to the model prediction results.

10. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the screen brightness adjustment method as described in any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the screen brightness adjustment method as described in any one of claims 1-8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the screen brightness adjustment method as described in any one of claims 1-8.

Images