A balanced control method, system and storage medium for partitioned backlight source

Abstract

The present invention provides a balanced control method, system and storage medium for partitioned backlight sources, which relates to the technical field of backlight source control. The method involves performing user data acquisition through an image acquisition device, establishing real-time user data, establishing real-time user needs based on display modes and real-time user data, capturing video streams in real time, performing partition aggregation of backlight sources based on picture analysis results, locating key features of partition aggregation results based on continuous gaze perception results, calling a joint control network, and establishing a joint partition backlight source control scheme based on key features, partition aggregation results and real-time user needs to perform display management of display screens. The method solves the technical problems in the prior art that the backlight source adjustment is not precise enough, the dynamic response is insufficient, resulting in uneven display effects and poor user viewing experience. The method achieves the technical effect of improving the accuracy and real-time response capability of backlight adjustment through partitioned backlight sources, and enhancing the uniformity of display effects and user viewing experience.

Description

Technical Field

[0001] The present invention relates to the technical field of backlight source control, and in particular to a balanced control method, system and storage medium for a partitioned backlight source. Background Art

[0002] In the current partition backlight control technology, due to the large adjustment granularity of the backlight, it is often impossible to achieve precise control of each subtle area of ​​the screen, resulting in uneven display effects in brightness and color. In addition, the existing technology has limited dynamic response capabilities when dealing with rapidly changing display content, and cannot adjust the backlight in time according to the picture requirements, thus affecting the smoothness and contrast of the picture. These problems not only reduce the visual effect of the display, but also seriously affect the user's viewing experience.

[0003] The existing technology has technical problems such as insufficient fine adjustment of the backlight source and insufficient dynamic response, which leads to uneven display effect and poor user viewing experience. Summary of the invention

[0004] The present application provides a balanced control method, system and storage medium for a partitioned backlight source, which are used to solve the technical problems in the prior art that the backlight source adjustment is not precise enough and the dynamic response is insufficient, resulting in uneven display effect and poor user viewing experience.

[0005] In view of the above problems, the present application provides a balanced control method, system and storage medium for a partitioned backlight source.

[0006] The first aspect of the present application provides a balanced control method for a partitioned backlight source, the method comprising: after user authorization, performing user data acquisition through an image acquisition device to establish real-time user data, the real-time user data including user facial data and body movement data; obtaining a display mode set by the user, and establishing real-time user needs based on the display mode and the real-time user data, wherein the display mode is a preset mode of the user, and the real-time user needs represent the fitting needs of the current user in the viewing state; capturing a video stream in real time, performing image analysis based on the video stream, performing partition aggregation of the backlight source based on the image analysis result, and establishing a partition aggregation result; performing continuous gaze perception on the user, locating key features of the partition aggregation result based on the continuous gaze perception result, calling a joint control network, and using the joint control network to establish a joint partition backlight source control scheme according to the key features, the partition aggregation result and the real-time user needs; and performing display management of the display screen through the joint partition backlight source control scheme.

[0007] According to a second aspect of the present application, a balanced control system for a partitioned backlight source is provided, the system comprising: a data acquisition module, the data acquisition module being used to perform user data acquisition through an image acquisition device after user authorization and permission, and establish real-time user data, the real-time user data including user facial data and body movement data; a user demand establishment module, the user demand establishment module being used to obtain a display mode set by the user, and establish real-time user demand based on the display mode and real-time user data, wherein the display mode is a preset mode of the user, and the real-time user demand represents the fitting demand of the current user in the viewing state; a partition aggregation result establishment module, the partition aggregation result establishment module being used to capture a video stream in real time, perform image analysis based on the video stream, perform partition aggregation of the backlight source based on the image analysis result, and establish a partition aggregation result; a control scheme establishment module, the control scheme establishment module being used to perform continuous gaze perception on the user, locate key features of the partition aggregation result based on the continuous gaze perception result, and call a joint control network to establish a joint partition backlight source control scheme based on the key features, the partition aggregation result and the real-time user demand by the joint control network; a display management module, the display management module being used to perform display management of the display screen through the joint partition backlight source control scheme.

[0008] According to a third aspect disclosed in the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any step of the first aspect disclosed in the present application is implemented.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The method provided in the embodiment of the present application is to perform user data collection through an image acquisition device after the user's authorization and permission, establish real-time user data, obtain the display mode set by the user, establish real-time user needs based on the display mode and real-time user data, wherein the display mode is the user's preset mode, and the real-time user needs represent the fitting needs of the current user in the viewing state, capture the video stream in real time, perform picture analysis based on the video stream, perform partition aggregation of the backlight source based on the picture analysis result, and establish the partition aggregation result. Perform continuous gaze perception on the user, locate the key features of the partition aggregation result based on the continuous gaze perception result, call the joint control network, and use the joint control network to establish a joint partition backlight control scheme based on the key features, partition aggregation results and real-time user needs, and perform display management of the display screen through the joint partition backlight control scheme. The technical effect of improving the accuracy and real-time response capability of backlight adjustment through partitioned backlight sources, and enhancing the uniformity of display effects and user viewing experience is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic diagram of a flow chart of a balanced control method for a partitioned backlight source provided in the present application;

[0013] Figure 2 A schematic diagram of the structure of a balanced control system for a partitioned backlight source provided in this application.

[0014] Explanation of the reference numerals: data collection module 11 , user demand establishment module 12 , partition aggregation result establishment module 13 , control scheme establishment module 14 , display management module 15 . DETAILED DESCRIPTION

[0015] The present application provides a balanced control method, system and storage medium for a partitioned backlight source, which is used to solve the technical problems in the prior art that the backlight source adjustment is not precise enough, the dynamic response is insufficient, and the display effect is uneven and the user viewing experience is poor. The technical effect of improving the accuracy and real-time response capability of backlight adjustment by partitioning the backlight source, and enhancing the uniformity of the display effect and the user viewing experience is achieved.

[0016] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.

[0017] Embodiment 1, as Figure 1 As shown, the present application provides a balanced control method for a partitioned backlight source, the method comprising:

[0018] After the user authorizes and permits, user data collection is performed through the image collection device to establish real-time user data, which includes user facial data and body movement data.

[0019] Specifically, the data collection process is started on the premise of obtaining explicit authorization from the user, and image acquisition devices such as cameras or special sensors are used to collect user data, record the user's facial features and body movements in real time, and establish a comprehensive real-time user data. The real-time user data reflects the data set of the user's current viewing status, including user facial data and body movement data. User facial data includes the user's facial expression, sight direction, facial position and emotional characteristics, etc., which are used to determine the user's current focus area and viewing emotional state. Body movement data involves the movement of the user's hands, head and other body parts, which is used to identify whether the user has interactive actions or actions that affect the viewing experience, such as head deflection or gesture indication, to determine the user's current activity status and possible light requirements. By comprehensively using image acquisition technology for real-time data collection, the backlight and content display of the display screen can be dynamically adjusted according to the user's viewing habits and postures, realizing the accurate capture of the user's personalized needs and providing reliable data support for the subsequent balanced control of the partition backlight source.

[0020] Obtain a display mode set by a user, and establish real-time user needs based on the display mode and real-time user data, wherein the display mode is a preset mode of the user, and the real-time user needs represent the fitting needs of the current user in the viewing state.

[0021] Specifically, the display mode is obtained from the user configuration, and the display mode is a mode pre-set by the user according to personal preferences or specific scenarios. For example, the display mode may include a cinema mode, a game mode, or a reading mode, and each mode corresponds to different brightness, contrast, and color settings. The preset display mode is cross-analyzed with the real-time user data obtained in real time from the image acquisition device. By analyzing the display model and the real-time user data, it is possible to determine whether the user's current viewing behavior conforms to the preset viewing mode, such as whether the user is concentrating on watching a movie or doing other activities. Based on the analysis results, real-time user needs are generated, and the real-time user needs represent the fitting needs of the current user's viewing state, that is, the user's specific needs for the current viewing experience. For example, if the user's facial data shows signs of fatigue, the real-time user needs may require adjusting the brightness or color temperature of the display mode to reduce visual fatigue. Based on the display mode and real-time user data, real-time user needs are established, and the backlight and screen display can be dynamically adjusted according to the user's current behavior and preferences, thereby providing users with a personalized and dynamic viewing experience.

[0022] Capture video streams in real time, perform image analysis based on the video streams, perform partition aggregation of backlight sources based on the image analysis results, and establish partition aggregation results.

[0023] Specifically, the video stream being played is obtained in real time in the form of continuous frames through image capture technology synchronized with the display screen. The video stream is captured at a certain frame rate (frames per second) to ensure the smoothness and real-time nature of the picture. Then, the video stream is decomposed into single-frame images, and each frame in the video stream is analyzed and processed. The picture analysis includes color recognition, edge detection, object tracking, etc. For example, color space conversion and threshold segmentation technology are used to identify different color areas in the picture, and edge detection algorithms are used to identify the outline and shape of the object. Then, the backlight source is partitioned and aggregated according to the picture analysis results. The partition aggregation is to divide the display screen backlight into multiple independently controllable areas according to the picture analysis results, so that the brightness of different areas can be adjusted independently according to content requirements. For example, a bright scene may need to increase the backlight of the area, while the darker part maintains a low brightness to enhance the overall contrast and detail clarity. Finally, according to the partition aggregation of the backlight source, the backlight source area after the partition aggregation is encoded and stored, and the partition aggregation result is established. The partition aggregation result includes parameters such as the position, size, and brightness of each partition. By capturing video streams in real time, performing image analysis, performing partition aggregation of backlight sources, and establishing partition aggregation results, precise control of the backlight source can be achieved, thereby optimizing display effects and improving user experience.

[0024] Furthermore, the method of performing picture analysis based on the video stream, performing partition aggregation of the backlight source based on the picture analysis result, and establishing the partition aggregation result also includes: transcoding the video stream, and performing image segmentation frame by frame to establish the image segmentation result; performing object recognition on the image segmentation result by using a deep learning model to establish picture key elements; establishing frame association of the video stream, performing dynamic content motion recognition based on the picture key elements and frame association, and generating pixel motion vectors; performing linkage aggregation according to the picture key elements and pixel motion vectors to establish the partition aggregation result.

[0025] Specifically, after obtaining the real-time video stream, first use transcoding tools or software, such as HandBrake, VLC, MediaCoder and other tools, to transcode the video stream, convert the original video stream into a format and encoding suitable for processing, and ensure that the video content can be efficiently decomposed and analyzed. Then, process the video stream frame by frame, and use image segmentation techniques, such as edge detection and segmentation algorithms, to decompose each frame of the image into different regions or objects to identify the boundaries and content structures of each screen element. After the image segmentation is completed, the deep learning model is used to perform object recognition on these segmentation results to generate a list of key elements of the screen. The identification of key elements is used to determine which areas require special backlight adjustment to improve the visual effect of the display. Deep learning models, such as convolutional neural networks, are obtained through training with a large amount of image data. During the training process, the deep learning model analyzes and learns multiple features in the image, such as edges, textures, shapes, and more complex patterns. With the continuous increase of training data and the continuous optimization of the model, the deep learning model can gradually improve the recognition accuracy of key elements in the image, such as faces, objects, backgrounds, etc. Through this process, the deep learning model can eventually accurately identify and classify the key elements in the picture. Next, the connection between the continuous frames of the video stream is analyzed to find the position and state changes of the same object in different frames, and to establish the association between the frames of the video stream. The dynamic content movement in the video is identified by using the frame association information and the key elements of the picture, and the position changes of the key elements in the continuous frames are tracked by calculating the picture similarity between the frames. By calculating the displacement vector between the pixels, a pixel point motion vector is generated, which refers to the displacement and speed information of each pixel in the video in the time series. Finally, the video screen is divided according to the key elements of the picture and the pixel point motion vector, and the areas with similar motion characteristics are linked and aggregated according to the movement trend and position of the elements to generate partition aggregation results. The partition aggregation results reflect the dynamic content structure in the video. For example, pixels with the same movement direction and speed are classified into one area, and these areas are further merged into larger partitions to achieve in-depth understanding and analysis of the video content and improve the accuracy and reliability of the balanced control of the partition backlight source.

[0026] Furthermore, the method of establishing frame association of video stream, identifying dynamic content motion based on key elements of the picture and frame association, and generating pixel motion vectors also includes: extracting the data structure of key elements of the picture, including position, category, execution degree, and timestamp; calculating the picture similarity of each time frame of the video stream, and generating frame association based on the picture similarity calculation results; identifying the motion position change and speed change of consecutive frames based on the data structure and frame association, and establishing pixel motion vectors.

[0027] Specifically, the data structure of the key elements of the picture is extracted through image recognition technology. The data structure includes information such as the position, category, execution degree and timestamp of the key elements. The position indicates the specific coordinates of the element in the picture, which is used to accurately locate the area where the element is located. The category is used to identify the type of the element, such as face, object or background, to help distinguish and classify different picture components. The execution degree refers to the importance or priority of the element in the current task. The timestamp is the recording time of each frame, which is used to trace the corresponding relationship between the change of the key element and time. Then, the picture similarity is calculated frame by frame on the video stream through feature matching algorithm or feature point-based matching, and repeated or similar content in adjacent frames is identified. The similarity and changes between consecutive frames are analyzed to generate frame association. Frame association is used to indicate the mapping relationship of how elements are associated and changed in video frames. Frame association helps to understand the dynamic content in the video stream and track the position and state changes of picture elements in consecutive frames. Finally, based on the data structure and frame association information, the motion position change and speed change of key elements in continuous frames are identified. Optionally, by matching key elements of the same category in adjacent frames, the movement trajectory of the elements in the picture is obtained. According to the position information of the key elements in continuous frames, their displacement is calculated, and then combined with the timestamp information, the speed change of the key elements, that is, the change in displacement per unit time, is obtained. The motion vector of each pixel is calculated based on the displacement and position change of the key elements in the time series. The pixel motion vector records the displacement path and speed information of the pixel in time, which can reflect the motion state of the object in the picture. Through identification and analysis, the tracking of the behavior of dynamic elements in the video stream is realized, providing data support for the subsequent partition backlight adjustment, ensuring accurate backlight control and display optimization in fast-motion scenes or dynamic pictures.

[0028] The user's continuous gaze perception is performed, and the key features of the partition aggregation results are located based on the continuous gaze perception results. The joint control network is called to establish a joint partition backlight control solution based on the key features, partition aggregation results and real-time user needs.

[0029] Specifically, an image acquisition device and high-precision eye tracking technology are used to capture and record the user's eye movement trajectory in real time. The user's gaze area during viewing is located by analyzing the user's continuous gaze perception. Continuous gaze perception determines the screen elements that the user focuses on for a long time by detecting the user's eye movement and line of sight. Based on the gaze perception results, such as the position of the gaze point, the moving path, etc., the key features of the corresponding partition aggregation results, that is, the main object or area of ​​the user's gaze in the displayed content, are located. Next, a joint control network is called, which is a complex neural network structure for coordinating and optimizing backlight adjustment, and comprehensively considers multiple factors, including but not limited to key features, partition aggregation results, and real-time user needs. The joint control network receives key features, partition aggregation results, and real-time user needs, and comprehensively processes these data to generate a joint partition backlight control scheme. The joint partition backlight control scheme not only takes into account the user's gaze behavior and needs, but also takes into account the uniformity of screen display. By dynamically adjusting the backlight brightness and color of different partitions of the display screen, accurate and personalized display optimization is achieved, thereby improving the user's viewing experience and ensuring that the area where the user is looking at gets the appropriate brightness and display balance.

[0030] Furthermore, the method of locating the key features of the partition aggregation results based on the continuous gaze perception results also includes: obtaining the temporal perception subject of the continuous gaze perception results, and performing frame picture backtracking based on the temporal perception subject, and generating a temporal subject identifier based on the frame picture backtracking results; performing subject interest authentication according to the temporal subject identifier and the temporal perception subject, and establishing a subject interest value; and locating the key features through the subject interest value and the partition aggregation results.

[0031] Specifically, the temporal perception subject of the continuous gaze perception result is obtained. The temporal perception subject refers to the different objects or areas that the user's sight focuses on during the continuous gaze process, which are arranged in chronological order to form a dynamic trajectory of gaze behavior, which is obtained through eye tracking technology and timestamp recording. Then, using the frame picture backtracking technology, according to the timestamp information of the temporal perception subject, backtrack to the corresponding frame picture, and obtain the user's gaze behavior and sight line changes in a specific time period through frame picture backtracking, that is, obtain the specific position, size and other information of each temporal perception subject on the screen, and generate a temporal subject identification. The temporal subject identification includes the number of subjects gazed by the user and their changes in different frames. Through the temporal subject identification, the dynamic changes of the user's sight line and the attention of different objects or areas during the continuous gaze process can be clearly understood. Then, according to the temporal subject identification and the temporal perception subject, subject interest authentication is performed. Subject interest authentication is to evaluate the user's interest in each temporal perception subject by analyzing the user's gaze behavior, such as gaze duration and gaze frequency. In this process, by considering factors such as the duration, frequency, and movement trajectory of the gaze point, the interest value of each temporal perception subject, namely the subject interest value, is comprehensively calculated. The higher the subject interest value, the higher the user's attention to the subject. Finally, the subject interest value and the partition aggregation results are used to locate the key features in the picture. That is, in the partition aggregation results, the partitions that the user is most concerned about are screened and located according to the subject interest value. These partitions are regarded as the locations of key features and are used for the formulation of subsequent backlight control strategies. By combining the subject interest value and the partition aggregation results, the area of ​​the picture that the user is most interested in can be more accurately located and identified, thereby achieving precise dynamic control of the backlight brightness, making the screen content more in line with the user's visual needs, and improving the uniformity of the display effect and the user's visual experience.

[0032] Furthermore, the subject interest authentication based on the time series subject identifier and the time series perception subject and the establishment of the subject interest value also includes: if the time series subject identifier is multiple subject identifiers, a perception order is established based on the time series perception subject and a subject priority is established; a time penalty factor is set, a time backtracking penalty is performed on the time series perception subject and the subject priority based on the time penalty factor, and a first interest value is established based on the time backtracking penalty result; an interest change trend is fitted for the time series perception subject based on the time series subject identifier, and a second interest value is established based on the interest change trend fitting result; and the subject interest authentication is completed based on the first interest value and the second interest value, and a subject interest value is established.

[0033] Specifically, if there are multiple temporal subject identifiers, that is, the user focuses on multiple screen elements at the same time, the perception order is established according to the order of these temporal perception subjects, so as to determine the priority of each subject in the user's attention. The perception order refers to the order in which the temporal perception subject interacts with the subject, and the subject priority is the ranking assigned to each subject after comprehensive evaluation based on the perception order. As time goes by, the user's interest may change or weaken. A time penalty factor is set to consider the influence of time when analyzing the user's gaze behavior. When performing time backtracking, the temporal perception subject and the subject priority are punished accordingly according to the time penalty factor, and the weight of the subject that the user paid attention to earlier but was subsequently ignored is weakened. And according to the time backtracking penalty result, the first interest value is calculated, which reflects the intensity and stability of the user's interest in the subject in the past period of time. At the same time, the interest change trend of the temporal perception subject is fitted based on the temporal subject identifier. The interest change trend fitting is an algorithm that predicts or identifies the trend of the subject's interest over time based on the interaction data between the temporal perception subject and the subject. By fitting the changes in user interests over time, an interest change trend is generated, that is, by analyzing the movement and retention of the user's sight in continuous frames, the growth or decline trend of interest is identified. A second interest value is established based on the fitting result, and the second interest change trend reflects the user's long-term interest fluctuations in different subjects in the picture. Finally, based on the first interest value and the second interest value, the historical interest intensity, stability and change trend of the subject are comprehensively considered to obtain the subject interest value. The subject interest value is a comprehensive value that can more comprehensively and accurately reflect the user's interest in each subject in the current video content, and then locate the key features that need to be adjusted most, and optimize the backlight source control to make the display more in line with the user's viewing preferences and improve the user experience.

[0034] Furthermore, the method of establishing a joint partitioned backlight source control scheme based on key features, partition aggregation results and real-time user needs using a joint control network also includes: mapping based on the key features and partition aggregation results based on the preprocessing layer of the joint control network to establish a mapping result; calling the reconstruction layer of the joint control network to establish a backlight demand reconstruction based on the mapping result and real-time user needs; synchronizing the backlight demand reconstruction result to the hidden layer of the joint control network, executing response scheme fitting, and establishing a joint partitioned backlight source control scheme based on the response scheme fitting result.

[0035] Specifically, the joint control network is obtained through training data, and its training data is a large amount of data content including user gaze behavior, video content features and backlight adjustment. The data set covers a variety of viewing scenarios and user behaviors to ensure that the network can cope with different situations in actual applications. The joint control network is a neural network architecture, including an input layer, a preprocessing layer, a reconstruction layer, a hidden layer, and an output layer. It can receive a variety of input data, such as key features of the picture, partition aggregation results, real-time user needs, and output a backlight control solution. The network is trained using the collected data. During the training process, the joint control network learns how to map the input to the best backlight adjustment solution. A supervised learning method is used to adjust the network through input / output samples, and a loss function is used to optimize the model parameters until the network model reaches a convergence state. Based on this, a joint control network that can respond to user needs in real time and optimize backlight adjustment is formed. The input layer of the joint control network receives key features, partition aggregation results, and real-time user demand information, and then inputs the key features and partition aggregation results into the preprocessing layer of the joint control network. The preprocessing layer performs preliminary analysis and processing on the input key features and partition aggregation results, and generates mapping results, i.e., backlight adjustment parameters corresponding to the input features. The mapping results are used to determine how each display partition should respond to the user's attention and the brightness requirements of the video content in the current screen. Next, the reconstruction layer of the joint control network is called. The reconstruction layer receives the mapping results and combines them with real-time user needs, i.e., viewing preferences and gaze behaviors obtained through image acquisition and user setting mode analysis, to further optimize the backlight adjustment strategy and generate backlight demand reconstruction. The backlight demand reconstruction reflects the user's current viewing state and the combination of screen features. Then, the backlight demand reconstruction result is synchronized to the hidden layer of the joint control network. The hidden layer performs response scheme fitting to analyze the relationship between the backlight demand and the screen response, ensuring that the output scheme can maintain consistency between dynamic picture changes and user behaviors, thereby ensuring smoother brightness changes and more uniform display effects. The response scheme refers to the specific control strategy for adjusting the backlight of the display screen fitted by the hidden layer of the joint control network based on the mapping results of the preprocessing layer and the backlight demand reconstruction results of the reconstruction layer, including dynamic and precise adjustment of the backlight brightness, color temperature and contrast of the display screen. By processing the response scheme fitting results, a joint partition backlight control scheme is generated and output through the output layer. The joint partition backlight control scheme will accurately control the backlight of each partition to match the real-time video content and the user's focus, thereby providing a more comfortable and personalized visual experience.

[0036] Display management of the display screen is carried out through a joint partition backlight control scheme.

[0037] Specifically, after obtaining the joint partition backlight control scheme, it is applied to the display screen to manage the display screen. By independently adjusting the backlight partitions of the screen, a high degree of fit between the picture content and the user's viewing needs is achieved. The joint partition backlight control scheme dynamically adjusts the brightness of each partition of the display screen based on the partition aggregation results and key features analyzed in real time, so that different areas respond independently according to the brightness requirements of the current picture content. For example, the darker parts of the display will reduce the backlight intensity to enhance contrast and details, while the brighter parts will increase the brightness to ensure the overall clarity of the picture. Display management through the joint partition backlight control scheme can make each partition transition smoothly according to the changes in the picture and user gaze, achieve precise backlight control, ensure that the screen brightness is adjusted quickly and the brightness is balanced when the scene changes, provide a smooth visual experience, and improve the comfort of the user's viewing experience.

[0038] Furthermore, the method also includes: establishing a user profile, and jointly recording the joint partition backlight control scheme and user viewing feedback; self-optimizing the joint partition backlight control scheme based on the joint recording results, and updating the self-optimization results to the user profile for executing subsequent viewing management of the user.

[0039] Specifically, a user profile is established to record the user's viewing habits, preference settings and usage history. The execution process of the joint partition backlight control scheme is jointly recorded with the user's viewing feedback in different scenarios, such as gaze time, gaze area and user's behavioral response, including the user's adjustment of brightness, color, contrast and other aspects and the response to the adjustment. Based on the joint recording results, the joint partition backlight control scheme is self-optimized, that is, by analyzing the user's viewing feedback during use and the execution effect of the joint partition backlight control scheme, the correlation and regularity are found. For example, if it is jointly recorded that the user tends to increase the brightness of the central area of ​​the screen to enhance the visual effect of the central figure of the picture, and reduce the brightness of the surrounding area to reduce interference when watching a movie, the joint partition backlight control scheme is automatically adjusted so that when the user watches similar content in the future, the screen can automatically distribute the brightness according to the user's preference. This self-optimization process is continuously iterated, so that the joint partition backlight scheme can better match the actual needs of the user. After the optimization is completed, the self-optimization results are updated to the user profile to ensure that the user is provided with the display effect that best fits his or her preferences in subsequent viewing, so as to achieve a more personalized and comfortable viewing experience.

[0040] Embodiment 2 is based on the same inventive concept as the balanced control method of a partitioned backlight source in the above embodiment. Figure 2 As shown, the present application provides a balanced control system for a partitioned backlight source, wherein the system includes:

[0041] The data acquisition module 11 is used to perform user data acquisition through an image acquisition device after the user authorizes and permits, and to establish real-time user data, wherein the real-time user data includes user facial data and body movement data.

[0042] The user demand establishment module 12 is used to obtain the display mode set by the user, and establish real-time user demand based on the display mode and real-time user data, wherein the display mode is a preset mode of the user, and the real-time user demand represents the fitting demand under the current user's viewing state.

[0043] The partition aggregation result establishing module 13 is used to capture the video stream in real time, perform picture analysis based on the video stream, perform partition aggregation of the backlight source based on the picture analysis result, and establish the partition aggregation result.

[0044] The control scheme establishing module 14 is used to perform continuous gaze perception of the user, locate the key features of the partition aggregation results based on the continuous gaze perception results, and call the joint control network to establish a joint partition backlight control scheme based on the key features, partition aggregation results and real-time user needs.

[0045] The display management module 15 is used to perform display management of the display screen through a joint partition backlight source control solution.

[0046] Furthermore, the partition aggregation result establishment module 13 is used to perform the following steps: transcode the video stream, and perform image segmentation frame by frame to establish image segmentation results; use a deep learning model to perform object recognition on the image segmentation results to establish key elements of the picture; establish frame associations of the video stream, perform dynamic content motion recognition based on key elements of the picture and frame associations, and generate pixel motion vectors; perform linkage aggregation based on key elements of the picture and pixel motion vectors to establish partition aggregation results.

[0047] Furthermore, the partition aggregation result establishment module 13 is also used to perform the following steps: extract the data structure of key elements of the picture, including position, category, execution degree, and timestamp; calculate the picture similarity of the video stream for each time frame, and generate frame associations based on the picture similarity calculation results; identify the motion position changes and speed changes of consecutive frames based on the data structure and frame associations, and establish pixel motion vectors.

[0048] Furthermore, the control scheme establishment module 14 is used to perform the following steps: obtain a temporal perception subject of the continuous gaze perception result, and perform frame picture backtracking based on the temporal perception subject, and generate a temporal subject identification based on the frame picture backtracking result; perform subject interest authentication according to the temporal subject identification and the temporal perception subject, and establish a subject interest value; locate key features through the subject interest value and the partition aggregation result.

[0049] Furthermore, the control scheme establishment module 14 is also used to perform the following steps: if the time series subject identifier is multiple subject identifiers, then a perception order is established based on the time series perception subject, and a subject priority is established; a time penalty factor is set, and a time backtracking penalty is performed on the time series perception subject and the subject priority based on the time penalty factor, and a first interest value is established based on the time backtracking penalty result; an interest change trend is fitted for the time series perception subject based on the time series subject identifier, and a second interest value is established based on the interest change trend fitting result; subject interest authentication is completed based on the first interest value and the second interest value, and a subject interest value is established.

[0050] Furthermore, the control scheme establishing module 14 is also used to perform the following steps: mapping based on the key features and partition aggregation results based on the preprocessing layer of the joint control network to establish a mapping result; calling the reconstruction layer of the joint control network to establish a backlight demand reconstruction based on the mapping result and real-time user needs; synchronizing the backlight demand reconstruction result to the hidden layer of the joint control network, executing response scheme fitting, and establishing a joint partition backlight source control scheme based on the response scheme fitting result.

[0051] Furthermore, the system is also used to perform the following steps: establish a user profile, and jointly record the joint partition backlight control scheme and user viewing feedback; self-optimize the joint partition backlight control scheme based on the joint recording results, and update the self-optimization results to the user profile for executing subsequent viewing management of the user.

[0052] Embodiment three, based on the same inventive concept as a balanced control method of a partitioned backlight source in the aforementioned embodiment, the present application also provides a storage medium, on which a computer program is stored, and when the computer program is executed, the steps of any one of the methods described in the aforementioned embodiment one are implemented.

[0053] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0054] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A balanced control method for a partitioned backlight source, characterized in that: The method comprises: After the user authorizes and permits, the image acquisition device is used to collect user data and establish real-time user data, wherein the real-time user data includes user facial data and body movement data; Acquire a display mode set by a user, and establish real-time user needs based on the display mode and real-time user data, wherein the display mode is a preset mode of the user, and the real-time user needs represent the fitting needs of the current user in the viewing state; Capture video streams in real time, perform image analysis based on the video streams, perform partition aggregation of backlight sources based on the image analysis results, and establish partition aggregation results; Perform continuous gaze perception on the user, locate the key features of the partition aggregation results based on the continuous gaze perception results, call the joint control network, and use the joint control network to establish a joint partition backlight control solution based on the key features, partition aggregation results and real-time user needs; Display management of the display screen through a joint partition backlight control solution; The performing of picture analysis based on the video stream, performing partition aggregation of the backlight source based on the picture analysis result, and establishing the partition aggregation result also includes: Transcode the video stream, perform image segmentation frame by frame, and establish image segmentation results; Use deep learning models to identify objects in image segmentation results and establish key elements of the picture; Establish frame association of video stream, perform dynamic content motion recognition based on key elements and frame association, and generate pixel motion vectors; Perform linkage aggregation based on key elements of the picture and pixel motion vectors to establish partition aggregation results; The step of establishing frame association of the video stream, performing dynamic content motion recognition based on key elements of the picture and frame association, and generating pixel motion vectors also includes: Extract the data structure of key elements of the screen, including location, category, execution degree, and timestamp; Calculate the picture similarity of each time frame of the video stream, and generate frame association based on the picture similarity calculation results; Based on the data structure and frame association, the motion position and speed changes of consecutive frames are identified and the pixel motion vector is established.

2. A balanced control method for a partitioned backlight source as claimed in claim 1, characterized in that: The key features of the positioning partition aggregation results based on the continuous gaze perception results also include: Acquire a temporal perception subject of the continuous gaze perception result, perform frame picture backtracking based on the temporal perception subject, and generate a temporal subject identifier based on the frame picture backtracking result; According to the time series subject identification and the time series perception subject, the subject interest authentication is performed to establish the subject interest value; Key features are located through subject interest values ​​and partition aggregation results.

3. A balanced control method for a partitioned backlight source as claimed in claim 2, characterized in that: The subject interest authentication is performed according to the time sequence subject identifier and the time sequence perception subject to establish the subject interest value, and further includes: If the time series subject identifier is multiple subject identifiers, a perception order is established based on the time series perception subject, and a subject priority is established; Setting a time penalty factor, performing a time backtracking penalty on the time-series perception subject and the subject priority based on the time penalty factor, and establishing a first interest value based on the time backtracking penalty result; Fitting the interest change trend of the time series perception subject based on the time series subject identifier, and establishing a second interest value based on the interest change trend fitting result; The subject interest authentication is completed according to the first interest value and the second interest value, and the subject interest value is established.

4. The balanced control method of a partitioned backlight source according to claim 1, characterized in that: The joint control network is used to establish a joint partition backlight source control solution according to key features, partition aggregation results and real-time user needs, and further includes: The preprocessing layer based on the joint control network performs mapping based on key features and partition aggregation results to establish mapping results; Call the reconstruction layer of the joint control network to establish backlight demand reconstruction based on the mapping results and real-time user needs; The backlight demand reconstruction result is synchronized to the hidden layer of the joint control network, and the response scheme fitting is performed. Based on the response scheme fitting result, a joint partition backlight source control scheme is established.

5. The balanced control method of a partitioned backlight source according to claim 1, characterized in that: The method further comprises: Establish user profiles and jointly record the joint partition backlight control scheme and user viewing feedback; The joint partition backlight control scheme is self-optimized based on the joint recording results, and the self-optimization results are updated to the user profile for subsequent viewing management of the user.

6. A balanced control system for a partitioned backlight source, characterized in that: The steps for implementing the method according to any one of claims 1 to 5 include: A data acquisition module, wherein the data acquisition module is used to perform user data acquisition through an image acquisition device after the user authorizes and permits, and establish real-time user data, wherein the real-time user data includes user facial data and body movement data; A user demand establishment module, the user demand establishment module is used to obtain a display mode set by a user, and establish real-time user demand based on the display mode and real-time user data, wherein the display mode is a preset mode of the user, and the real-time user demand represents the fitting demand of the current user in the viewing state; A partition aggregation result establishment module, wherein the partition aggregation result establishment module is used to capture the video stream in real time, perform picture analysis based on the video stream, perform partition aggregation of the backlight source based on the picture analysis result, and establish the partition aggregation result; A control scheme establishing module, wherein the control scheme establishing module is used to perform continuous gaze perception of the user, locate key features of the partition aggregation results based on the continuous gaze perception results, and call a joint control network to establish a joint partition backlight control scheme based on the key features, the partition aggregation results and real-time user needs; A display management module is used to perform display management of a display screen through a joint partition backlight source control solution.

7. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a balanced control method for a partitioned backlight source according to any one of claims 1 to 5 are implemented.

Images