Image content brightness regulation method for teaching display device and related device

Abstract

The application discloses a kind of image content brightness regulation methods and related equipment for teaching display device, effectively avoid the situation that some areas are too bright or too dark, improve the definition of key information, to improve the quality of teaching.The present application includes: real-time acquisition sampling information, sampling information includes the image of teaching display device output, the ambient light parameter of teaching display device and the brightness feedback information of user;Based on semantic segmentation model, the image in sampling information is segmented into text area, chart area and background area to obtain structured segmentation result;According to the feature parameters of text area, chart area and background area are extracted according to structured segmentation result;Structured segmentation result, feature parameters, ambient light parameters and brightness feedback information are input into layered reinforcement learning model to obtain screen partition brightness regulation parameters;Sampling information and brightness regulation parameters are mapped and superimposed to obtain superposition result.

Description

Technical Field

[0001] This application relates to the field of teaching aids technology, and in particular to a method and related equipment for adjusting the brightness of image content in teaching display devices. Background Technology

[0002] In teaching settings, LED screens are crucial display devices, and their proper brightness control plays a vital role in enhancing teaching effectiveness. Existing methods for LED screen brightness control mostly rely on simple adjustments using ambient light sensors, or only provide a uniform adjustment to the overall screen brightness. These methods cannot intelligently adjust brightness based on the specific image content displayed. When displaying complex images or videos, this can easily lead to areas being too bright or too dark, making key information difficult to see and impacting teaching quality. Summary of the Invention

[0003] To address the aforementioned technical problems, this application provides a method and related equipment for adjusting the brightness of image content in a teaching display device.

[0004] The technical solution provided in this application is described below:

[0005] The first aspect of this application provides a method for adjusting the brightness of image content in a teaching display device, comprising:

[0006] Real-time acquisition of sampling information, including images output by the teaching display device, ambient light parameters of the teaching display device, and brightness feedback information from the user;

[0007] Based on the semantic segmentation model, the image in the sampled information is segmented into text regions, chart regions, and background regions to obtain structured segmentation results. The structured segmentation results include the coordinate range and area ratio of the text regions, chart regions, and background regions.

[0008] Feature parameters of the text region, the chart region, and the background region are extracted based on the structured segmentation results;

[0009] The structured segmentation result, the feature parameters, the ambient light parameters, and the brightness feedback information are input into the hierarchical reinforcement learning model. The hierarchical reinforcement learning model is used to initialize the brightness strategies of the text region, the chart region, and the background region based on the structured segmentation result, then integrate the brightness feedback information to adjust the brightness weights of the sub-regions, and calculate the brightness benchmark value according to the ambient light parameters to obtain the screen partition brightness control parameters.

[0010] The sampling information and the brightness control parameters are mapped and superimposed to obtain the superimposed result;

[0011] Based on the superposition result, screen control instructions are generated;

[0012] The screen control command is sent to the teaching display device to control the image display and brightness.

[0013] Optionally, based on a semantic segmentation model, the image in the sampled information is segmented into text regions, chart regions, and background regions to obtain structured segmentation results, including:

[0014] The images in the sampled information are preprocessed to obtain preprocessed images, which are adapted to the input requirements of the semantic segmentation model;

[0015] The preprocessed image is input into the semantic segmentation model, which performs classification prediction on each pixel in the preprocessed image and outputs the original segmentation result of the region category to which each pixel belongs;

[0016] After smoothing the edges and merging adjacent similar regions on the original segmentation result, a structured segmentation result is obtained. The structured segmentation result includes the coordinate range and area ratio of the text region, chart region, and background region.

[0017] Optionally, feature parameters of the text region, the chart region, and the background region are extracted, including:

[0018] The pixel range and boundaries of each region are determined based on the structured segmentation results.

[0019] Based on the pixel range and boundaries of each region, extract the features of text size, font contrast and character density in the text region, extract the features of color level, line complexity and element distribution density in the chart region, and extract the features of color uniformity, texture features and brightness mean in the background region.

[0020] The extracted feature parameters are normalized to obtain the feature parameters of the text region, the chart region, and the background region.

[0021] Optionally, the sampling information and the brightness control parameters are mapped and superimposed to obtain a superimposed result, including:

[0022] Based on the coordinate range of the background region in the structured segmentation result, the text region, chart region, and background region of the image in the sampled information are associated with the screen partition brightness control parameters;

[0023] The brightness adjustment parameters associated with the text area, chart area, and background area are mapped to each pixel in the text area, chart area, and background area respectively, to form a correspondence between pixels and brightness.

[0024] The superposition result is generated based on the correspondence between pixels and brightness, and the superposition result includes the brightness adjustment parameters corresponding to each pixel.

[0025] Optionally, screen control instructions are generated based on the superposition result, including:

[0026] Based on the overlay results, extract the coordinate ranges and corresponding target brightness values ​​of the text region, chart region, and background region;

[0027] The coordinate range and target brightness value formats are converted into a target format, which is a communication protocol format supported by the teaching display device, and an instruction sequence that the teaching display device can parse is generated.

[0028] Screen control instructions are generated based on the instruction sequence.

[0029] Optionally, sending the screen control command to the teaching display device to control image display and brightness includes:

[0030] The screen control commands are encapsulated according to a communication protocol and sent to the teaching display device via a communication link;

[0031] The system receives the execution results of the instructions returned by the teaching display device and verifies whether the brightness adjustment is effective as expected. If an execution error occurs, a retry mechanism is triggered until the teaching display device completes the image display and brightness adjustment.

[0032] Optionally, after acquiring sampling information in real time, including the image output by the teaching display device, the ambient light parameters of the teaching display device, and the user's brightness feedback information, the method further includes:

[0033] The sampling information is preprocessed to obtain preprocessed sampling information.

[0034] A second aspect of this application provides an image content brightness adjustment system for a teaching display device, comprising:

[0035] The first acquisition unit is used to acquire sampling information in real time, including the image output by the teaching display device, the ambient light parameters of the teaching display device, and the brightness feedback information from the user.

[0036] The second acquisition unit performs text region, chart region, and background region segmentation on the image in the sampled information based on the semantic segmentation model to obtain structured segmentation results. The structured segmentation results include the coordinate range and area ratio of the text region, chart region, and background region.

[0037] An extraction unit is used to extract feature parameters of the text region, the chart region, and the background region based on the structured segmentation results;

[0038] The input unit is used to input the structured segmentation result, the feature parameters, the ambient light parameters, and the brightness feedback information into the hierarchical reinforcement learning model. The hierarchical reinforcement learning model is used to initialize the brightness strategies of the text region, the chart region, and the background region based on the structured segmentation result, then integrate the brightness feedback information to adjust the brightness weights of the sub-regions, and calculate the brightness benchmark value according to the ambient light parameters to obtain the screen partition brightness control parameters.

[0039] The overlay unit is used to map and overlay the sampling information and the brightness control parameters to obtain an overlay result;

[0040] A generation unit is used to generate screen control instructions based on the superposition result;

[0041] The control unit is used to send the screen control command to the teaching display device to realize the control of image display and brightness.

[0042] A third aspect of this application provides an image content brightness adjustment system for a teaching display device, the system comprising:

[0043] Processor, memory, input / output units, and bus;

[0044] The processor is connected to the memory, the input / output unit, and the bus;

[0045] The memory stores a program, which the processor invokes to perform the method as described in the first aspect and any one of the first aspects.

[0046] A fourth aspect of this application provides a computer-readable storage medium on which a program is stored, which, when executed on a computer, performs the method as described in the first aspect and any one of the first aspects.

[0047] As can be seen from the above technical solutions, this application has the following beneficial effects:

[0048] The above-mentioned technical content can achieve the following technical effects:

[0049] 1. This application utilizes a semantic segmentation model to accurately segment the text region, chart region, and background region of the image output by the teaching display device, obtains the coordinate range and area ratio of each region, and then extracts the feature parameters of each region based on these structured segmentation results. Based on these parameters, the brightness strategies of different regions are initialized, thereby enabling brightness optimization according to the characteristics of different content in the image, ensuring that the text region is clear and readable, the chart region is brightly colored, and the background region is not too bright and distracting, thus achieving brightness adjustment that meets actual needs.

[0050] 2. This application uses a hierarchical reinforcement learning model to generate brightness control parameters. It can not only initialize the brightness strategy based on the structured segmentation results, but also integrate brightness feedback information to adjust the brightness weights of different regions. Through learning and optimization, it can gradually master the complex relationship between different image content, ambient light conditions and user feedback, thereby generating more reasonable and effective screen partition brightness control parameters and improving the intelligence level of brightness control.

[0051] 3. This application, through image segmentation, feature extraction, model calculation, and result overlay, finally generates and sends screen control commands, realizing automated image display and brightness control without frequent manual intervention. It can quickly and efficiently adapt to different teaching scenarios and user needs, improving the utilization efficiency of teaching display equipment.

[0052] 4. This application can intelligently adjust the brightness according to the specific image content displayed on the screen. When displaying complex images or videos, it can effectively avoid some areas being too bright or too dark, improve the clarity of key information, and thus improve the quality of teaching. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a schematic diagram of an embodiment of the image content brightness adjustment method for teaching display devices according to this application;

[0055] Figure 2 This is a schematic diagram of another embodiment of the image content brightness adjustment method for teaching display devices according to this application;

[0056] Figure 3 This is a schematic diagram of another embodiment of the image content brightness adjustment method for teaching display devices according to this application;

[0057] Figure 4 This is a schematic diagram of another embodiment of the image content brightness adjustment method for teaching display devices according to this application;

[0058] Figure 5 This is a schematic diagram of another embodiment of the image content brightness adjustment method for teaching display devices according to this application;

[0059] Figure 6This is a schematic diagram of another embodiment of the image content brightness adjustment method for teaching display devices according to this application;

[0060] Figure 7 This is a schematic diagram of an embodiment of the image content brightness adjustment system for teaching display devices according to this application;

[0061] Figure 8 This is a schematic diagram of another embodiment of the image content brightness adjustment system for educational display devices according to this application. Detailed Implementation

[0062] In teaching settings, LED screens are crucial display devices, and their proper brightness control plays a vital role in enhancing teaching effectiveness. Existing methods for LED screen brightness control mostly rely on simple adjustments using ambient light sensors, or only provide a uniform adjustment to the overall screen brightness. These methods cannot intelligently adjust brightness based on the specific image content displayed. When displaying complex images or videos, this can easily lead to areas being too bright or too dark, making key information difficult to see and impacting teaching quality.

[0063] Based on this, this application provides a method and related equipment for adjusting the brightness of image content in teaching display devices. It can intelligently adjust the brightness according to the specific image content displayed on the screen, effectively avoiding overly bright or dark areas when displaying complex images or videos, improving the clarity of key information, and thus improving teaching quality.

[0064] Please see Figure 1 The first aspect of this application provides a method for adjusting the brightness of image content in a teaching display device, the method comprising:

[0065] 101. Real-time acquisition of sampling information, including images output by the teaching display device, ambient light parameters of the teaching display device, and brightness feedback information from the user;

[0066] 102. Based on a semantic segmentation model, the image in the sampled information is segmented into text regions, chart regions, and background regions to obtain structured segmentation results, wherein the structured segmentation results include the coordinate range and area ratio of the text regions, chart regions, and background regions;

[0067] 103. Extract feature parameters of the text region, the chart region, and the background region based on the structured segmentation results;

[0068] 104. Input the structured segmentation result, the feature parameters, the ambient light parameters, and the brightness feedback information into the hierarchical reinforcement learning model. The hierarchical reinforcement learning model is used to initialize the brightness strategies of the text region, the chart region, and the background region based on the structured segmentation result, then integrate the brightness feedback information to adjust the brightness weights of the sub-regions, and calculate the brightness benchmark value according to the ambient light parameters to obtain the screen partition brightness control parameters.

[0069] 105. Map and superimpose the sampling information and the brightness control parameters to obtain the superposition result;

[0070] 106. Generate screen control instructions based on the superposition result;

[0071] 107. Send the screen control command to the teaching display device to realize the control of image display and brightness.

[0072] In this embodiment, sampling information is first acquired in real time. This sampling information includes the image output by the teaching display device, the ambient light parameters of the teaching display device, and the user's brightness feedback information. Then, based on a semantic segmentation model, the image in the sampling information is segmented into text regions, chart regions, and background regions to obtain structured segmentation results. The structured segmentation results include the coordinate range and area ratio of the text regions, chart regions, and background regions. Next, feature parameters of the text regions, chart regions, and background regions are extracted based on the structured segmentation results. Then, the structured segmentation results, feature parameters, ambient light parameters, and brightness feedback information are input into a hierarchical reinforcement learning model. The hierarchical reinforcement learning model is used to initialize the brightness strategies of the text regions, chart regions, and background regions based on the structured segmentation results. Then, it integrates the brightness feedback information to adjust the brightness weights of the sub-regions and calculates the brightness baseline value based on the ambient light parameters to obtain the screen partition brightness control parameters. The sampling information and brightness control parameters are mapped and superimposed to obtain the superposition result. Based on the superposition result, a screen control command is generated and sent to the teaching display device to realize the control of image display and brightness.

[0073] In step 101, sampling information is first acquired in real time. This sampling information includes three key categories: The first category is the image output by the teaching display device. This data requires capturing the entire screen currently displayed on the device in real time, including the playing courseware pages, the written blackboard content, and the video footage being presented. The second category is the ambient light parameters of the teaching display device. This is achieved by using the device's built-in light sensor to detect data such as the ambient light intensity and color temperature in real time, such as the strong light near the classroom window during the day, the weak light indoors in the evening, or the local light changes after the lights are turned on. The third category is the user's brightness feedback information. This information comes from the user's feedback on brightness, such as when a student or teacher clicks the "brighter" or "darker" feedback command on the device's interactive interface. This information reflects the user's subjective perception of the current brightness. Therefore, after acquiring the sampling information in real time, step 102 is executed.

[0074] In step 102, the image in the sampled information is then segmented into regions based on the semantic segmentation model. This step is crucial for transforming the original image into structured data, because only by clearly defining the regional distribution of different content in the image can targeted brightness adjustment be achieved. First, a pre-trained semantic segmentation model needs to be invoked. It should be noted that this model has been trained on a large amount of teaching scenario image data and can accurately identify text regions in the image, such as title text, body text, formula symbols, etc. in courseware; chart regions, such as line charts, bar charts, flowcharts, diagrams, etc.; and background regions, such as blank backgrounds, solid color backgrounds, or irrelevant decorative elements in courseware.

[0075] During the segmentation process, the model classifies each pixel in the image, determines the type of region it belongs to, and then generates a structured segmentation result. The structured segmentation result not only includes the coordinate range of each region in the image, such as the pixel coordinates of the upper left and lower right corners of the text region, clarifying the specific position of the region on the screen, but also calculates the area proportion of each region. For example, in a certain courseware page, the text region accounts for 40%, the chart region accounts for 35%, and the background region accounts for 25%. These data will directly provide a basis for subsequent extraction of feature parameters and initialization of brightness strategies.

[0076] In step 103, feature parameters for each region are extracted based on the structured segmentation results. Specifically, for text regions, the feature parameters to be extracted include font size (e.g., 16pt Songti, 24pt Heiti), contrast between text color and background color (e.g., contrast of black text on a white background, contrast of dark text on a light background), and text density (e.g., whether text is densely arranged on a slide, whether there are large blocks of text or scattered short sentences). These parameters determine the required brightness threshold for text regions to ensure that users can clearly recognize the text. For chart regions, feature parameters include chart color saturation (e.g., the vibrancy of colors in a color chart, whether there are low-saturation light-colored charts), line thickness of chart elements (e.g., line width of a line chart, border thickness of a bar chart), and size of data labels in the chart (e.g., the size of numbers on the chart axes, the size of labels next to data points). These parameters affect the recognizability of the chart content and need to be matched with corresponding brightness. For background regions, the main feature parameters are the background color type (e.g., solid color background, gradient background, background with slight texture), the color difference between the background and foreground content, and the color difference between text and charts.

[0077] In step 104, the structured segmentation results, feature parameters, ambient light parameters, and brightness feedback information are then input into the hierarchical reinforcement learning model to generate brightness control parameters. Specifically, the hierarchical reinforcement learning model first initializes the brightness strategy for each region based on the structured segmentation results: for example, based on the area proportion of the text region, if the text region accounts for a large proportion, such as exceeding 50%, the brightness baseline of the text region is prioritized during initialization; if the chart region accounts for an even larger proportion, such as exceeding 40%, the initial brightness weight of the chart region is appropriately increased to ensure the display effect of the core content.

[0078] Next, the model will integrate brightness feedback information to adjust the brightness weights of different areas: if the user reports that "the text is not clear", the brightness weight of the text area will be increased, and the brightness ratio of the text area will be appropriately increased; if the user reports that "the screen is too bright overall", the overall brightness coefficient will be reduced based on the brightness weights of each area, while prioritizing ensuring that the brightness of key areas such as text and charts is not lower than the minimum clarity threshold.

[0079] Finally, the model calculates a baseline brightness value based on ambient light parameters. For example, when the ambient light intensity is high, such as when a classroom is brightly lit during the day, the overall baseline brightness value is increased to prevent the screen from appearing dim due to excessive ambient light. When the ambient light intensity is low, such as when only dim lighting is on indoors at night, the baseline brightness value is decreased to prevent the screen from being too bright and glaring. Simultaneously, based on the characteristic parameters of each area, the brightness of each area is fine-tuned based on the baseline value. For example, if the color saturation of a chart area is low, the brightness will be appropriately increased above the baseline value to ensure the chart is clear. Ultimately, these calculations yield the screen's zoned brightness control parameters, which specify the exact brightness values ​​or adjustment ranges for text, charts, and background.

[0080] In step 105, the sampling information and brightness control parameters are then mapped and superimposed. The purpose of this step is to map the abstract control parameters to the specific display content, preparing for the generation of device-executable instructions. Specifically, firstly, a mapping relationship is established between the image regions in the sampling information and the brightness control parameters: based on the coordinate range of each region in the structured segmentation results, the brightness values ​​of the corresponding regions in the brightness control parameters are precisely matched to the specific locations in the image. For example, the coordinate range of the text region (x1, y1) to (x2, y2) corresponds to the superimposed brightness value L1 of the text region in the brightness control parameters; the coordinate range of the chart region (x3, y3) to (x4, y4) corresponds to the superimposed brightness value L2; and the coordinate range of the background region (x5, y5) to (x6, y6) corresponds to the superimposed brightness value L3.

[0081] During the overlay process, the image content details in the sampling information are also taken into account. For example, in the text area, the brightness of text of different font sizes may be slightly adjusted based on L1. The brightness of the text area with smaller font size is slightly higher than that of the text area with larger font size to ensure that all text is clear. In the chart area, the brightness of the thinner lines will also be finely adjusted based on L2 to avoid the lines from being blurry due to insufficient brightness. Finally, an overlay result containing the corresponding brightness values ​​of each pixel area is formed. This result presents the brightness state that needs to be adjusted to each position on the screen.

[0082] In step 106, screen control commands are generated based on the superposition result. The purpose of this step is to convert the superimposed brightness data into a command format that the teaching display device can recognize and execute. Specifically, the brightness information in the superposition result is first parsed, and the coordinate range and corresponding brightness value of each area are converted into parameters that the device can understand. For example, pixel coordinates are converted into the physical coordinates of the device screen, such as a coordinate system with the upper left corner of the screen as the origin, the horizontal axis as the X-axis, and the vertical axis as the Y-axis. The brightness values ​​are converted into brightness levels that the device's backlight module can recognize, such as a brightness value range of 0-255.

[0083] Next, the structural framework of the generated instruction is determined. The instruction needs to include the instruction type, such as "zone brightness adjustment"; the adjustment time parameter, such as "execute immediately" or "execute after 0.5 seconds delay," to ensure real-time performance; the physical coordinates of each area and its corresponding brightness level; and verification information to ensure the instruction is not tampered with or damaged during transmission. For example, an adjustment instruction might be: "Zone brightness adjustment, execute immediately: text area (100,200)-(800,600) brightness 180, chart area (200,700)-(700,900) brightness 200, background area (0,0)-(1000,1000) brightness 120, verification code XXX." Such an instruction allows the device to clearly understand the brightness adjustment required for each area.

[0084] In step 107, finally, the screen control command is sent to the teaching display device to achieve image display and brightness control. Specifically, the generated control command is first transmitted to the control chip of the teaching display device through the device communication module. After receiving the command, the control chip first verifies the command to confirm its completeness and correctness. If the verification passes, it parses the coordinates and brightness parameters of each area in the command, and then sends a control signal to the backlight driver module of the teaching display device. According to the control signal, the backlight driver module adjusts the brightness of the backlight LEDs in the corresponding areas. For example, the backlight LEDs corresponding to the text area are adjusted to a brightness level of 180, those corresponding to the chart area are adjusted to 200, and those corresponding to the background area are adjusted to 120, thereby achieving precise control of the brightness of each area of ​​the screen.

[0085] Meanwhile, the device will provide real-time feedback on the adjusted status. If subsequent sampling information shows that the current brightness still needs optimization, it will re-enter the first sampling process and start a new round of adjustment to ensure that the screen brightness always matches the display content and environmental requirements, providing users with a good visual experience.

[0086] Please refer to Figure 2 According to some embodiments of the present invention, step 102, which involves segmenting the image in the sampled information into text regions, chart regions, and background regions based on a semantic segmentation model to obtain structured segmentation results, may specifically include, but is not limited to, the following:

[0087] 201. Preprocess the image in the sampled information to obtain a preprocessed image, wherein the preprocessed image is adapted to the input requirements of the semantic segmentation model;

[0088] 202. Input the preprocessed image into the semantic segmentation model, the semantic segmentation model performs classification prediction on each pixel in the preprocessed image, and outputs the original segmentation result of the region category to which each pixel belongs;

[0089] 203. After smoothing the edges and merging adjacent similar regions on the original segmentation result, a structured segmentation result is obtained. The structured segmentation result includes the coordinate range and area ratio of the text region, chart region, and background region.

[0090] In this embodiment, after acquiring the image from the sampling information, it is preprocessed to obtain a preprocessed image that meets the input requirements of the semantic segmentation model. This step ensures that the image can be effectively recognized and processed by the model. After preprocessing, the preprocessed image is input into the semantic segmentation model. The model classifies and predicts each pixel in the image and outputs the original segmentation result of the region category to which each pixel belongs. The semantic segmentation model is built on a deep convolutional neural network and, after being trained on a large amount of teaching image data labeled with text, charts, and background regions, has the ability to recognize the features of different regions. During processing, the model extracts features from the preprocessed image through multiple convolutional layers, gradually transitioning from low-level edge and texture features to high-level semantic features, such as the stroke features of text, the line and color block features of charts, and the uniformity features of the background. Subsequently, the model classifies each pixel, determining whether it belongs to a text region, a chart region, or a background region, and assigns a corresponding category label to each pixel, for example, using 0 to represent a background region, 1 to represent a text region, and 2 to represent a chart region. This results in the original segmentation result, which is essentially a label matrix of the same size as the preprocessed image. Each matrix element corresponds to the category of a pixel in the image, and it fully records the preliminary classification of all pixels in the image.

[0091] Next, the original segmentation results need to be processed by smoothing edges and merging adjacent regions of the same type to obtain a structured segmentation result containing the coordinate range and area proportion of each region. Edge smoothing primarily aims to optimize the roughness of region boundaries. Through morphological operations or Gaussian smoothing, jagged or burr-like edges are eliminated, making the boundaries of text, chart, and background regions smoother and more natural. For example, scattered misclassified pixels at the edges of text regions in the original segmentation result can be corrected to the correct category through smoothing, making the outline of the text region more consistent with its actual shape. Merging adjacent regions of the same type addresses the possibility of scattered small regions of the same category in the original segmentation result. These small regions may be due to model prediction errors, such as the chart region being incorrectly segmented into multiple small blocks. By calculating the category similarity and distance of adjacent regions, adjacent small regions belonging to the same category are merged into a complete region, ensuring the integrity and coherence of each region.

[0092] After edge smoothing and region merging, the coordinate range of each region can be determined. By traversing the pixels of each region, the pixel coordinates of its top-left and bottom-right corners are found, thus clarifying the specific location of the region in the image. Simultaneously, the number of pixels contained in each region is counted, and combined with the total number of pixels in the image, the area proportion of each region is calculated. For example, the proportion of pixels in a text region to the total number of pixels is its area proportion. This structured segmentation result not only accurately reflects the distribution of different content in the image, but its coordinate range and area proportion data can also directly provide a basis for subsequent feature parameter extraction and brightness initialization strategies.

[0093] Please refer to Figure 3 According to some embodiments of the present invention, the feature parameters for extracting the text region, the chart region, and the background region in step 103 may specifically include, but are not limited to, the following:

[0094] 301. Determine the pixel range and boundary of each region based on the structured segmentation results;

[0095] 302. Based on the pixel range and boundaries of each region, extract the features of text size, font contrast and character density in the text region, extract the features of color level, line complexity and element distribution density in the chart region, and extract the features of color uniformity, texture features and brightness mean in the background region.

[0096] 303. Normalize the extracted feature parameters to obtain the feature parameters of the text region, the chart region, and the background region.

[0097] In this embodiment, the pixel range and boundaries of each region are determined based on the structured segmentation results. This step involves fine-grained region delineation based on the coordinate ranges and area proportions of the text, chart, and background regions. The coordinate ranges given by the structured segmentation results are usually the approximate boundaries of each region. For example, a text region may be represented by a rectangle, but the actual text distribution may not fill the entire rectangle. Therefore, pixel-level analysis is needed to determine the precise pixel range of each region, i.e., which specific pixels belong to the text region and which belong to the chart or background region. Simultaneously, the precise boundaries between regions must be clearly defined, such as the pixels at the boundary between text and background, and the pixels distinguishing chart edges from the background. This process utilizes information such as the color and grayscale values ​​of image pixels, and uses thresholding methods to determine the affiliation of each pixel, ultimately forming an accurate pixel set and clear boundary lines for each region.

[0098] Next, different features are extracted based on the pixel range and boundaries of each region. For text regions, features such as text size, font contrast, and character density are extracted: Text size extraction requires first identifying the pixel range of a single character, which is determined by calculating the height or width of the character's bounding rectangle. For example, if a character occupies 20 pixels vertically, it corresponds to the font size. Font contrast is calculated by determining the difference in grayscale values ​​between text pixels and surrounding background pixels. For example, the grayscale value of black text pixels is 0, while that of a white background is 255; the contrast is represented by the difference or ratio between the two. Character density is the number of characters per unit area. Combining this with the pixel range of the text region, the number of characters per square centimeter is calculated to reflect the density of the text.

[0099] For the chart area, extract the features of color hierarchy, line complexity, and element distribution density: Color hierarchy is determined by analyzing the types and transitions of different colors within the chart area. For example, a chart containing 10 different colors with natural transitions has richer color hierarchy than a chart containing only 2 colors; Line complexity requires identifying various types of lines in the chart and measuring them by calculating parameters such as the total length of the lines and the number of turns. The longer the lines and the more turns, the higher the complexity; Element distribution density is the number of various elements in the chart, such as data points, bars, and legends, per unit area, reflecting the density of the elements.

[0100] For the background area, features such as color uniformity, texture features, and average brightness are extracted: color uniformity is determined by calculating the variance of pixel color values ​​in the background area; the smaller the variance, the more uniform the color. Texture features are determined by analyzing the spatial distribution of pixel gray values ​​to determine whether the background is smooth and textureless or has subtle grids or gradient textures. Average brightness is calculated as the average brightness value of all pixels in the background area, reflecting the overall brightness of the background.

[0101] Finally, the extracted feature parameters are normalized. Because different feature parameters have significantly different units and numerical ranges—for example, text size might be expressed in pixels (10-50 pixels), while font contrast might be expressed as a percentage (30%-90%)—directly using these feature parameters can affect the accuracy and efficiency of subsequent model calculations. Normalization transforms these feature parameters to a uniform numerical range, typically 0-1 or -1-1. For instance, the pixel values ​​for text size are scaled proportionally to the 0-1 range, with a maximum value of 50 pixels corresponding to 1 and a minimum value of 10 pixels corresponding to 0; the percentage of font contrast is directly converted to a decimal between 0 and 1. This process eliminates the influence of different units on the feature parameters, ensuring they have equal weight and comparability when inputting into the hierarchical reinforcement learning model. This guarantees that the model can more accurately utilize these feature parameters to calculate and adjust brightness strategies, resulting in more reasonable screen partition brightness control parameters.

[0102] Please refer to Figure 4 According to some embodiments of the present invention, step 105, which maps and superimposes the sampling information and the brightness adjustment parameters to obtain the superposition result, may specifically include, but is not limited to, the following:

[0103] 401. Based on the coordinate range of the background region in the structured segmentation result, associate the text region, chart region, and background region of the image in the sampling information with the screen partition brightness control parameters;

[0104] 402. Map the brightness adjustment parameters associated with the text area, chart area, and background area to each pixel in the text area, chart area, and background area respectively, to form a correspondence between pixels and brightness.

[0105] 403. Generate a superimposed result based on the correspondence between pixels and brightness, wherein the superimposed result includes the brightness adjustment parameters corresponding to each pixel.

[0106] In this embodiment, based on the coordinate range of the background region in the structured segmentation result, the text region, chart region, and background region of the image in the sampled information are associated with the screen partition brightness control parameters. The key here is to establish an overall spatial reference using the coordinate range of the background region, as the background region typically covers most of the screen, and its coordinate range provides a positioning benchmark for the text and chart regions. The structured segmentation result clearly defines the coordinate range of each of the three regions; for example, the text region might be (x1, y1) to (x2, y2), the chart region is (x3, y3) to (x4, y4), and the background region is (x5, y5) to (x6, y6). The screen partition brightness control parameters also correspond to the brightness values ​​of these three regions. For example, the brightness of the text region is L1, the brightness of the chart region is L2, and the brightness of the background region is L3. By using the coordinate range of the background region, the relative positions of the text and chart regions on the entire screen can be calibrated, ensuring that each region accurately matches the corresponding brightness control parameter and avoiding mismatches between regions and brightness parameters.

[0107] Next, the brightness control parameters associated with the text area, chart area, and background area are mapped to each pixel within their respective areas, establishing a correspondence between pixels and brightness. After establishing the association between areas and brightness parameters, further refinement is needed for each pixel. For the text area, the associated brightness parameter L1 is applied to every pixel within that area, whether it's the pixel containing the text strokes or the white space pixels surrounding the text; as long as it belongs to the coordinate range of the text area, it will correspond to brightness L1. Similarly, every pixel within the chart area, including the chart's lines, color blocks, and labels, is mapped to brightness parameter L2; all pixels within the background area correspond to brightness parameter L3. This process requires precise coordinate matching to ensure that each pixel can be accurately assigned to its respective area and given the corresponding brightness value. For example, if a pixel's coordinates (a, b) are determined to belong to the chart area, it will be associated with brightness L2, giving each pixel a clear brightness control target.

[0108] Then, based on the correspondence between pixels and brightness, a superimposed result is generated. This superimposed result contains the brightness adjustment parameters corresponding to each pixel. After mapping, the brightness value that each pixel on the screen needs to be adjusted to is clearly defined. Integrating this information forms the superimposed result. This result is essentially a brightness parameter matrix with the same resolution as the original image. Each element in the matrix corresponds to the brightness adjustment parameter of the pixel at the same position in the original image. For example, if the original image has a resolution of 1920×1080, the superimposed result is a 1920×1080 matrix, where the value at position (i,j) is the brightness value corresponding to pixel (i,j) in the original image. This superimposed result completely records the brightness adjustment scheme of the entire screen, including the brightness differences in different areas (text, charts, and background areas have different brightness), while ensuring the consistency of brightness within the same area. This ensures that the adjustment commands can accurately act on every position on the screen, achieving precise zoned brightness control.

[0109] Please refer to Figure 5 According to some embodiments of the present invention, the step 106 of generating screen control instructions based on the superposition result may specifically include, but is not limited to, the following:

[0110] 501. Extract the coordinate range and corresponding target brightness values ​​of the text area, chart area, and background area based on the overlay result;

[0111] 502. Convert the coordinate range and target brightness value format into a target format, wherein the target format is a communication protocol format supported by the teaching display device, and generate an instruction sequence that the teaching display device can parse;

[0112] 503. Generate screen control instructions according to the instruction sequence.

[0113] In this embodiment of the application, when extracting the coordinate range and corresponding target brightness value of the text area, chart area and background area based on the overlay result, it is necessary to parse the previously obtained overlay result. It should be noted that the overlay result already contains the mapping relationship between each area of ​​the screen and the corresponding brightness value. What needs to be done here is to accurately extract the specific information of the three key areas: text, chart and background.

[0114] For each region, it's not enough to determine its coordinate range on the screen—this range needs to be precise enough for the device to clearly identify the region's boundaries, such as a text region being a rectangle extending from the top-left corner (x1, y1) to the bottom-right corner (x2, y2). Furthermore, the target brightness value for that region must be extracted; this is the optimal brightness value determined after the preceding calculations. This extracted information will serve as the basis for subsequent instruction generation, ensuring that brightness adjustment for each region has a clear target and range.

[0115] Next, the coordinate range and target brightness value formats are converted to the target format, which is the communication protocol format supported by the teaching display device, and an instruction sequence that the device can parse is generated. Different teaching display devices typically use different communication protocols, which have specific requirements for data formats. For example, the coordinate representation may need to be converted from pixel coordinates to a device-defined physical coordinate system, and the brightness value may need to be converted from a numerical range of 0-255 to hexadecimal code or a specific numerical range specified by the device protocol. During the format conversion process, the device's communication protocol specifications must be followed to ensure that the converted coordinate range and target brightness value can be correctly recognized by the device.

[0116] After the format conversion is completed, the coordinate range and target brightness value of each region can be organized into an ordered sequence of instructions according to the instruction structure required by the protocol. Each instruction corresponds to the brightness adjustment task of a region. These instruction sequences need to clearly show the correspondence between the region order, coordinate information and brightness value.

[0117] Finally, screen control instructions are generated based on the instruction sequence. Specifically, the previously generated instruction sequence is integrated and encapsulated to form a complete instruction that conforms to the device's execution standards. First, the instruction sequence is checked for completeness to ensure that the control instructions for each area are included without omissions or errors. Then, according to the instruction format required by the device, necessary control information is added to the instruction sequence, such as the start and end identifiers of the instruction, and checksums. This information is used by the device to verify the validity and completeness of the instruction, preventing errors or tampering during transmission. Simultaneously, the execution method of the instruction is further considered, such as whether it is executed immediately or delayed; these execution parameters are also included in the final screen control instructions. After such integration and encapsulation, the generated screen control instructions can be directly received and parsed by the teaching display device, thereby accurately executing the brightness control operations for each area and optimizing the screen display effect.

[0118] Please refer to Figure 6 According to some embodiments of the present invention, sending the screen control command to the teaching display device in step 107 to realize the control of image display and brightness may specifically include, but is not limited to, the following:

[0119] 601. Encapsulate the screen control commands according to the communication protocol and send them to the teaching display device through the communication link;

[0120] 602. Receive the instruction execution result returned by the teaching display device and verify whether the brightness adjustment is effective as expected; if the execution is abnormal, trigger the retry mechanism until the teaching display device completes the image display and brightness adjustment.

[0121] In this embodiment, after generating the screen control commands, these commands need to be encapsulated according to the communication protocol supported by the teaching display device. This is a crucial step to ensure that the commands can be correctly recognized and received by the device. Different teaching display devices may use different communication protocols; for example, some devices use the TCP / IP protocol, some may use the RS-232 serial port protocol, and others may support the HDMI-CEC protocol. During the encapsulation process, the various contents of the control commands need to be structured according to the protocol format specified by the device, such as adding protocol header information, command length field, check field, and protocol tail identifier.

[0122] In this application, taking the TCP / IP protocol as an example, the encapsulated instructions are divided into data packets that conform to network transmission specifications. Each data packet contains the aforementioned protocol-related fields, ensuring that it can be accurately sent to the receiving port of the teaching display device when transmitted through wired or wireless communication links.

[0123] After the encapsulated command is sent out through the communication link, the next step is to wait for and receive the command execution result from the teaching display device. Upon receiving the command, the teaching display device parses and executes it, generating corresponding feedback information upon completion. This feedback information is also encapsulated and sent back according to the previous communication protocol. The receiving end, upon receiving this feedback result, needs to parse it, focusing on verifying whether the brightness adjustment has taken effect as expected. Specifically, it checks whether the actual brightness values ​​of each area in the feedback result are consistent with the target values ​​set in the adjustment command. For example, whether the text area has reached the specified brightness level, and whether the brightness adjustment of the chart area meets expectations. It also confirms whether the device has completed the brightness adjustment of all areas within the specified time.

[0124] If the verification result shows an execution anomaly, such as the actual brightness of a certain area deviating from the target value beyond the allowable range, or the device returning an execution failure status code, the system will immediately trigger a retry mechanism. The retry mechanism will first record the anomaly information, including the area where the anomaly occurred, the deviation value, and the error code, for subsequent analysis of the cause of the problem. Subsequently, the system will regenerate the screen control commands and re-encapsulate and send them according to the communication protocol. This retry process will continue until a successful execution result is received from the teaching display device, meaning that the brightness control of all areas has reached the expected target, thus ultimately achieving precise control of image display and brightness.

[0125] Please see Figure 7 A second aspect of this application provides an image content brightness adjustment system for a teaching display device, the system comprising:

[0126] The first acquisition unit 701 is used to acquire sampling information in real time, including the image output by the teaching display device, the ambient light parameters of the teaching display device, and the brightness feedback information of the user.

[0127] The second acquisition unit 702 performs text region, chart region and background region segmentation on the image in the sampled information based on the semantic segmentation model to obtain a structured segmentation result, wherein the structured segmentation result includes the coordinate range and area ratio of the text region, chart region and background region.

[0128] Extraction unit 703 is used to extract feature parameters of the text region, the chart region and the background region based on the structured segmentation result;

[0129] The input unit 704 is used to input the structured segmentation result, the feature parameters, the ambient light parameters and the brightness feedback information into the hierarchical reinforcement learning model. The hierarchical reinforcement learning model is used to initialize the brightness strategies of the text region, the chart region and the background region based on the structured segmentation result, then integrate the brightness feedback information to adjust the brightness weights of the sub-regions, and calculate the brightness reference value according to the ambient light parameters to obtain the screen partition brightness control parameters.

[0130] The superposition unit 705 is used to map and superimpose the sampling information and the brightness control parameters to obtain a superposition result;

[0131] The generation unit 706 is used to generate screen control instructions based on the superposition result;

[0132] The control unit 707 is used to send the screen control command to the teaching display device to realize the control of image display and brightness.

[0133] Please see Figure 8 This application also provides an image content brightness adjustment system for a teaching display device, the system comprising:

[0134] Processor 801, memory 802, input / output unit 803, bus 804;

[0135] The processor 801 is connected to the memory 802, the input / output unit 803, and the bus 804;

[0136] The memory 802 stores a program, and the processor 801 calls the program to execute any of the methods described above.

[0137] This application also relates to a computer-readable storage medium on which a program is stored, which, when run on a computer, causes the computer to perform any of the methods described above.

[0138] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0139] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0140] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0141] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0142] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for regulating the brightness of image content of a teaching display device, characterized in that, The method comprises the following steps: real-time acquisition of sampling information, the sampling information including an image output by a teaching display device, an ambient light parameter of the teaching display device, and brightness feedback information of a user; segmentation of the image in the sampling information into a text region, a chart region, and a background region based on a semantic segmentation model to obtain a structured segmentation result, the structured segmentation result including coordinate ranges and area proportions of the text region, the chart region, and the background region; determination of pixel ranges and boundaries of each region based on the structured segmentation result; extraction of features of text size, font contrast, and character density in the text region, extraction of features of color level, line complexity, and element distribution density in the chart region, and extraction of features of color uniformity, texture characteristics, and brightness average in the background region according to the pixel ranges and boundaries of each region; normalization of the extracted feature parameters to obtain feature parameters of the text region, the chart region, and the background region; input of the structured segmentation result, the feature parameters, the ambient light parameter, and the brightness feedback information into a hierarchical reinforcement learning model, the hierarchical reinforcement learning model being configured to initialize brightness strategies of the text region, the chart region, and the background region based on the structured segmentation result, then adjust regional brightness weights by fusing the brightness feedback information, and calculate a brightness reference value based on the ambient light parameter to obtain screen partition brightness control parameters; mapping and superimposition of the sampling information and the brightness control parameters to obtain a superimposition result; generation of a screen control instruction according to the superimposition result; sending of the screen control instruction to the teaching display device to realize image display and brightness control.

2. The method for teaching display device image content brightness regulation according to claim 1, characterized in that, Segmentation of the image in the sampling information into a text region, a chart region, and a background region based on a semantic segmentation model to obtain a structured segmentation result comprises the following steps: preprocessing of the image in the sampling information to obtain a preprocessed image, the preprocessed image being adapted to input requirements of the semantic segmentation model; input of the preprocessed image into the semantic segmentation model, the semantic segmentation model performing classification prediction on each pixel in the preprocessed image and outputting an original segmentation result of the region category to which each pixel belongs; obtaining of a structured segmentation result after edge smoothing and adjacent same region merging of the original segmentation result, the structured segmentation result including coordinate ranges and area proportions of the text region, the chart region, and the background region.

3. The method for teaching display device image content brightness regulation according to claim 1, characterized in that, Mapping and superimposition of the sampling information and the brightness control parameters to obtain a superimposition result comprises the following steps: association of the text region, the chart region, and the background region of the image in the sampling information with screen partition brightness control parameters based on the coordinate range of the background region in the structured segmentation result; mapping of the brightness control parameters associated with the text region, the chart region, and the background region to each pixel point in the text region, the chart region, and the background region respectively to form a corresponding relationship between pixels and brightness; generation of a superimposition result based on the corresponding relationship between pixels and brightness, the superimposition result including brightness control parameters corresponding to each pixel.

4. The method for teaching display device image content brightness regulation according to claim 1, characterized in that, The method comprises the following steps: According to the superposition result, a screen regulation instruction is generated, which comprises: According to the superposition result, the coordinate range and the corresponding target brightness value of the text area, the chart area and the background area are extracted; The format of the coordinate range and the target brightness value is converted into a target format, which is a communication protocol format supported by the teaching display device, and an instruction sequence that can be parsed by the teaching display device is generated; 5. The method for teaching display device image content brightness regulation according to claim 1, characterized in that, According to the instruction sequence, a screen regulation instruction is generated. The screen regulation instruction is sent to the teaching display device to realize the regulation of image display and brightness, which comprises: The screen regulation instruction is encapsulated according to the communication protocol and sent to the teaching display device through the communication link; 6. The method for teaching display device image content brightness regulation according to claim 1, wherein, The instruction execution result returned by the teaching display device is received to verify whether the brightness regulation is effective as expected; if the execution is abnormal, a retry mechanism is triggered until the teaching display device completes the regulation of image display and brightness. After the sampling information, including the image output by the teaching display device, the ambient light parameter of the teaching display device and the brightness feedback information of the user, is obtained in real time, the method further comprises:

7. A system for regulating the brightness of image content for a display device used for teaching, characterized by, The sampling information is preprocessed to obtain preprocessed sampling information. It comprises: A first acquisition unit is configured to acquire sampling information in real time, which comprises an image output by a teaching display device, an ambient light parameter of the teaching display device and brightness feedback information of a user; A second acquisition unit is configured to segment the image in the sampling information into a text area, a chart area and a background area based on a semantic segmentation model to obtain a structured segmentation result, which contains the coordinate range and the area ratio of the text area, the chart area and the background area; An extraction unit is configured to determine the pixel range and the boundary of each area based on the structured segmentation result; extract the features of the text size, the font contrast and the character density in the text area, the features of the color level, the line complexity and the element distribution density in the chart area, and the features of the color uniformity, the texture feature and the brightness mean value in the background area according to the pixel range and the boundary of each area; and normalize each type of feature parameter to obtain the feature parameters of the text area, the chart area and the background area; An input unit is configured to input the structured segmentation result, the feature parameters, the ambient light parameter and the brightness feedback information into a hierarchical reinforcement learning model, which is configured to initialize the brightness strategy of the text area, the chart area and the background area based on the structured segmentation result, then adjust the regional brightness weight by fusing the brightness feedback information, and calculate a brightness reference value according to the ambient light parameter to obtain a screen partition brightness regulation parameter; A superposition unit is configured to map and superimpose the sampling information and the brightness regulation parameter to obtain a superposition result; A generation unit is configured to generate a screen regulation instruction according to the superposition result; 8. A system for regulating the brightness of image content for a display device used for teaching, characterized by, A regulation unit is configured to send the screen regulation instruction to the teaching display device to realize the regulation of image display and brightness. The system comprises: A processor, a memory, an input / output unit and a bus; The processor is connected with the memory, the input / output unit and the bus; The memory stores a program, and the processor invokes the program to execute the method in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a program, and the program executes the method in any one of claims 1 to 6 when executed on the computer.

Images