A screen projection control method and an intelligent central control system for exhibition halls

Abstract

This invention provides a screen projection control method and an intelligent central control system for exhibition halls. The method includes: receiving a screen projection request sent by a user terminal and parsing the request; evaluating the display effect of each screen projection content on each exhibition hall screen based on the content information of the projection content and the exhibition hall screen information; calculating the evaluation value of each screen projection content under various evaluation indicators; weighted summing of the evaluation values ​​under various evaluation indicators to obtain a comprehensive evaluation value of each screen projection content on each exhibition hall screen; allocating exhibition hall screens to each screen projection content in descending order of comprehensive evaluation value; and controlling each screen projection content to be displayed on the corresponding exhibition hall screen. This allows for intelligent allocation of the most suitable exhibition hall screen to each screen projection content based on its characteristics and the display parameters of the exhibition hall screens, ensuring optimal display effect and improving the visual experience of screen projection.

Description

Technical Field

[0001] This invention relates to the field of smart exhibition hall technology, and in particular to a screen projection control method and an intelligent central control system for exhibition halls. Background Technology

[0002] With the widespread application of multimedia display technology, screen projection in exhibition halls has become an important method of information display. In existing exhibition hall screen projection scenarios, the allocation of projected content is usually based on simple rules, such as the order of projection requests or the screen's idle state. This method fails to fully consider the characteristics of the projected content and the display parameters of the exhibition hall screen, resulting in poor display quality. For example, if high-resolution images or videos are arbitrarily allocated to a low-resolution screen, problems such as blurry images and loss of detail will occur, affecting the viewer's visual experience. Summary of the Invention

[0003] This invention provides a projection control method and an intelligent central control system for exhibition halls, which improves projection effects through intelligent projection content allocation strategies, providing a better visual experience for viewers and also enhancing the operational efficiency of exhibition halls.

[0004] To solve the above problems, the present invention adopts the following technical solution:

[0005] This invention provides a screen mirroring control method, comprising:

[0006] Receive screen mirroring requests from users;

[0007] The screen mirroring request is parsed to obtain the parsed content, which includes multiple screen mirroring contents and the content information of each screen mirroring content;

[0008] Based on the content information and exhibition hall screen information, the display effect of each projection content on each exhibition hall screen is evaluated, and the adaptability of each projection content on each exhibition hall screen is calculated. The adaptability includes the evaluation value of the projection content under each evaluation index.

[0009] The evaluation values ​​of each projected content under each evaluation indicator are weighted and summed to obtain the comprehensive evaluation value of each projected content on each exhibition hall screen. Each projected content is assigned an exhibition hall screen in descending order of the comprehensive evaluation value, and each projected content is controlled to be projected and displayed on the corresponding exhibition hall screen.

[0010] Optionally, controlling the display of each projected content on the corresponding exhibition hall screen includes:

[0011] The data format of each screen-projected content is determined based on the content information;

[0012] Adjust the display style of the projected content according to the data format of each projected content and the display parameters of the corresponding exhibition hall screen;

[0013] The projected content is displayed on the corresponding exhibition hall screen according to the described display style.

[0014] Optionally, controlling the display of each projected content on the corresponding exhibition hall screen includes:

[0015] When the projected content is detected to be an image, the light intensity and ambient color temperature in the exhibition hall environment are detected.

[0016] The screen color temperature value of the exhibition hall screen when displaying pure white is detected, and a target correction color temperature value is set according to the ambient color temperature value.

[0017] Construct an optimal function model between the target corrected color temperature value and the screen color temperature value, and use the optimal function model to calculate the correction coefficients;

[0018] The saturation and brightness of the projected content are adjusted according to the light intensity and correction coefficient, and the color temperature value of the projected content is adjusted according to the correction coefficient to generate the first projected content.

[0019] Control the first projection content to be projected and displayed on the corresponding exhibition hall screen.

[0020] Optionally, controlling the display of each projected content on the corresponding exhibition hall screen includes:

[0021] When the screen-projected content is detected to be video, each video frame of the screen-projected content is divided into image blocks of a fixed size.

[0022] Find the target video frame that is most similar to each image block in the preset set of reference video frames, and calculate the motion vector of each target video frame;

[0023] The motion vector of each target video frame is scaled proportionally to form multiple intermediate frames;

[0024] Each intermediate frame is inserted between the corresponding video frames of the projected content to generate the second projected content;

[0025] Control the second projection content to be projected and displayed on the corresponding exhibition hall screen.

[0026] Optionally, controlling the display of each projected content on the corresponding exhibition hall screen includes:

[0027] Each of the projected content items is matched with the reference projected content in the content library;

[0028] When it is determined that a third type of projection content exists in the content library that matches the projection content and has superior content quality, the third type of projection content replaces the original projection content and is projected onto the corresponding exhibition hall screen; wherein, the generation method of the reference projection content includes:

[0029] The exhibit information and user tags of the exhibition hall are obtained, and the exhibit information and user tags are input into the trained projection content generation model to generate reference projection content for each exhibit under different user tags. The projection content generation model uses a convolutional neural network model as the underlying architecture.

[0030] Furthermore, before inputting the exhibit information and user tags into the trained projection content generation model, the process also includes:

[0031] Obtain multiple samples of screen-projected content;

[0032] Each of the screen projection content samples is labeled and assigned a sample tag, which includes the exhibit information introduced by each screen projection content sample and the matching user tag;

[0033] Training data is constructed based on the multiple screen projection content samples and their corresponding sample labels;

[0034] The convolutional neural network model is iteratively trained using the training data. After the convolutional neural network model converges, the trained convolutional neural network model is used as the screen projection content generation model.

[0035] Furthermore, after iteratively training the convolutional neural network model using the training data, the method further includes:

[0036] Calculate the loss value of the convolutional neural network model after training based on the cross-entropy loss function;

[0037] When the loss value is greater than a preset value, the learning rate of the convolutional neural network model is obtained, and a decay coefficient is set according to the loss value;

[0038] The learning rate of the convolutional neural network model is adjusted according to the decay coefficient, and the adjusted convolutional neural network model is retrained using the training data until the loss value is lower than the preset value, thus obtaining a trained screen projection content generation model.

[0039] Furthermore, after controlling each projected content to be displayed on the corresponding exhibition hall screen, the method further includes:

[0040] Each of the projected content items is matched with the reference projected content in the content library;

[0041] When it is determined that there is a fourth screen-casting content associated with the screen-casting content in the content library, the fourth screen-casting content is sent to the user terminal.

[0042] The system receives a confirmation instruction from the user terminal for the fourth projection content. After the projection of the original projection content corresponding to the fourth projection content is completed, the system controls the fourth projection content to be projected and displayed on the exhibition hall screen corresponding to the original projection content.

[0043] Optionally, parsing the screen mirroring request to obtain the parsed content includes:

[0044] The screen projection request is identified, and an identification result is generated. The identification result includes different content types and corresponding encoding formats in the screen projection request.

[0045] Based on the recognition results, the screen projection request is divided into multiple data blocks;

[0046] Each data block is parsed using a matching parsing algorithm, and the data blocks containing the target screen parameters are independently extracted and transformed to match the display parameters of the exhibition hall screen, thus obtaining the parsed content corresponding to each data block.

[0047] The present invention also provides an intelligent central control system for exhibition halls, comprising:

[0048] The receiving module is used to receive screen mirroring requests sent by the user client;

[0049] The parsing module is used to parse the screen projection request to obtain parsed content, which includes multiple screen projection contents and content information of each screen projection content;

[0050] The evaluation module is used to evaluate the display effect of each projection content on each exhibition hall screen based on the content information and exhibition hall screen information, and to calculate the adaptability of each projection content on each exhibition hall screen. The adaptability includes the evaluation value of the projection content under each evaluation index.

[0051] The projection display module is used to perform a weighted summation of the evaluation values ​​of each projection content under various evaluation indicators to obtain the comprehensive evaluation value of each projection content on each exhibition hall screen. The module then allocates exhibition hall screens to each projection content in descending order of the comprehensive evaluation values ​​and controls each projection content to be projected and displayed on the corresponding exhibition hall screen.

[0052] Compared with the prior art, the technical solution of the present invention has at least the following advantages:

[0053] This invention provides a screen projection control method and an intelligent central control system for exhibition halls. It receives screen projection requests from user terminals, parses these requests, and obtains multiple projection contents and their information. Then, based on the content information and exhibition hall screen information, it evaluates the display effect of each projection content on each exhibition hall screen, calculating the evaluation value of each content under various evaluation indicators. By weighted summing of the evaluation values ​​under each indicator, it obtains a comprehensive evaluation value for each content on each exhibition hall screen. Finally, it assigns exhibition hall screens to each projection content in descending order of comprehensive evaluation value and controls each content to be projected and displayed on its corresponding screen. This intelligently allocates the most suitable exhibition hall screen to each projection content based on its characteristics and the display parameters of the exhibition hall screens, ensuring optimal display effects and improving the visual experience. Simultaneously, it enables synchronous projection from multiple exhibition hall screens. Attached Figure Description

[0054] Figure 1 This is a flowchart of an embodiment of a screen projection control method according to the present invention;

[0055] Figure 2 This is a flowchart of another embodiment of the screen projection control method of the present invention;

[0056] Figure 3 This is a structural block diagram of one embodiment of the intelligent central control system for exhibition halls of the present invention. Detailed Implementation

[0057] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0058] Please refer to Figure 1 As shown, the present invention provides a screen projection control method, including:

[0059] S11. Receive the screen mirroring request sent by the user client;

[0060] S12. The screen projection request is parsed to obtain parsed content, which includes multiple screen projection contents and content information of each screen projection content;

[0061] S13. Evaluate the display effect of each projection content on each exhibition hall screen based on the content information and exhibition hall screen information, and calculate the adaptability of each projection content on each exhibition hall screen. The adaptability includes the evaluation value of the projection content under each evaluation index.

[0062] S14. The evaluation values ​​of each projection content under each evaluation index are weighted and summed to obtain the comprehensive evaluation value of each projection content on each exhibition hall screen. The exhibition hall screens are assigned to each projection content in descending order of the comprehensive evaluation values, and each projection content is controlled to be projected and displayed on the corresponding exhibition hall screen.

[0063] Visitors or exhibition hall staff can send a screen projection request to the exhibition hall's intelligent central control system via the network using their mobile devices (such as mobile phones, tablets, computers, etc.). This screen projection request is a data packet containing basic screen projection information.

[0064] The screen mirroring request can be a data packet in JSON or XML format, containing user identity information, identifiers of the mirrored content, and the type of the mirrored content. Furthermore, to ensure data transmission security, the screen mirroring request can be encrypted, such as using SSL / TLS encrypted communication.

[0065] After receiving a screen projection request from a user, the exhibition hall's intelligent central control system needs to parse the request data and extract the projection content and related information. For example, a JSON parser or XML parser can be used to process the projection request data, extracting attributes of each projection content, such as content type (image, video, text, etc.), resolution, encoding format, and duration. It can also extract device information from the user's device, such as device type, operating system version, and network status.

[0066] Based on the attributes of the projected content and the parameters of the exhibition hall screens, the display effect of the projected content on each screen is evaluated, and an adaptability score is calculated. For example, the degree of matching between the resolution of the projected content and the screen resolution is calculated; a higher adaptability score indicates a better display effect. The accuracy of color reproduction of the projected content on the screen is assessed; the screen's color gamut and color calibration affect this metric. The location and layout of the screens in the exhibition hall are considered, and the viewing angle and distance of the audience are evaluated; screens within the audience's line of sight have higher visibility. The current usage status of the screens is assessed to avoid allocating too much content to the same screen, which could lead to display lag or overload. Then, the evaluation metrics for each projected content on each screen are quantified to obtain an adaptability score.

[0067] Based on the comprehensive evaluation score of each projected content on each screen, the exhibition hall screens are allocated to the content in descending order of score, and the display is controlled. For example, the evaluation scores of each projected content under various evaluation indicators are weighted and summed to obtain a comprehensive evaluation score. The weights can be adjusted according to actual needs, such as setting a higher weight for display quality. Then, projected content is allocated starting with the screen with the highest comprehensive evaluation score. If that screen has already been allocated, the allocation is attempted to the next screen with the highest evaluation score. During the allocation process, the screen load is dynamically adjusted to ensure balanced utilization of each screen.

[0068] This invention provides a screen projection control method that receives screen projection requests from user terminals, parses these requests, and obtains multiple projection contents and their content information. Then, based on the content information of the projection contents and the information of the exhibition hall screens, the display effect of each projection content on each exhibition hall screen is evaluated, and an evaluation value for each projection content under various evaluation indicators is calculated. By weighted summing of the evaluation values ​​under each evaluation indicator, a comprehensive evaluation value for each projection content on each exhibition hall screen is obtained. Finally, exhibition hall screens are assigned to each projection content in descending order of comprehensive evaluation value, and each projection content is controlled to be displayed on its corresponding exhibition hall screen. This method intelligently allocates the most suitable exhibition hall screen to each projection content based on its characteristics and the display parameters of the exhibition hall screens, ensuring optimal display effects and improving the visual experience of screen projection. Simultaneously, it enables synchronous screen projection from multiple exhibition hall screens.

[0069] In one embodiment, controlling the projection of each content onto the corresponding exhibition hall screen includes:

[0070] The data format of each screen-projected content is determined based on the content information;

[0071] Adjust the display style of the projected content according to the data format of each projected content and the display parameters of the corresponding exhibition hall screen;

[0072] The projected content is displayed on the corresponding exhibition hall screen according to the described display style.

[0073] After the projected content is assigned to the corresponding exhibition hall screen, the system can determine the data format of the projected content. The data format refers to information such as the file type, encoding method, resolution, and color space of the projected content.

[0074] For example, projected content includes images, videos, and documents. Images can be in JPEG or PNG format; videos can be in MP4 or AVI format. For video content, its encoding method, such as H.264 or H.265, is also required. Documents can be in PDF, Word, or other formats. For image and video content, resolution is a key data format information. The system can know the original resolution of the projected content in order to adjust its display size on the exhibition hall screen later.

[0075] Secondly, different projected content can use different color spaces, such as RGB and YUV, and the color space determines how colors are represented. Furthermore, for video content, the video frame rate determines the smoothness of video playback, affecting the viewer's experience.

[0076] After determining the data format of the projected content and allocating screens in the exhibition hall, the system can adjust the display style of the projected content based on the display parameters of the exhibition hall screens (such as resolution, size, color temperature, brightness, etc.) and the data format of the projected content. Adjustments to the display style include image scaling, color correction, and video frame rate adaptation to ensure that the projected content is displayed optimally on the exhibition hall screens.

[0077] Specifically, if the projected content is an image, the system can scale the image according to the resolution of the exhibition hall screen. For example, when the original resolution of the image is higher than that of the exhibition hall screen, the system uses a high-quality downsampling algorithm to reduce the image size while minimizing the loss of image quality; when the original resolution of the image is lower than that of the exhibition hall screen, the system uses an upsampling algorithm to enlarge the image and enhance its clarity.

[0078] When scaling images, the system can maintain the image's aspect ratio, preventing stretching or compression distortion. If the aspect ratio of the exhibition hall screen differs from the original image's aspect ratio, it can be adjusted by cropping or padding. For example, when projecting an image with a 16:9 aspect ratio onto a screen with a 4:3 aspect ratio, the system can add black padding bars on both sides of the image to maintain its aspect ratio and integrity.

[0079] The color temperature of the exhibition hall screen affects the color performance of the projected content. The system can adjust the colors of the projected content based on the color temperature parameters of the exhibition hall screen. For example, if the color temperature of the exhibition hall screen is too cool (e.g., 7000K), while the projected content is designed according to a standard color temperature (e.g., 6500K), the system can correct the color temperature by reducing the blue component in the image, making the colors of the projected content more natural.

[0080] Furthermore, the color space of the projected content may differ from that of the exhibition hall screen. The system can perform color space conversion to ensure that the colors of the projected content are accurately reproduced on the exhibition hall screen. For example, it can convert video content in the YUV color space to the RGB color space required by the exhibition hall screen.

[0081] Considering the influence of ambient light in the exhibition hall, the system can utilize ambient light sensors to obtain color temperature and intensity information of the ambient light, and further adjust the colors of the projected content. In brighter environments, the contrast and saturation of the projected content are appropriately increased to make the image clearer and more vibrant; in darker environments, the contrast is appropriately reduced to avoid the image being too dark.

[0082] After adjusting the display style of the projected content, the system will send the adjusted content to the corresponding exhibition hall screen for display according to the predetermined display style, ensuring that the projected content can be presented to the audience accurately.

[0083] For example, network streaming protocols can be used to send the adjusted projection content to the exhibition hall screen. During transmission, the projection content can also be appropriately compressed to improve transmission efficiency while minimizing the impact of compression on image and video quality. For instance, for video content, efficient video coding standards (such as H.264) can be used for compression, and then decompressed and restored on the exhibition hall screen to improve the display effect of the projection content.

[0084] Please refer to Figure 2 As shown, in one embodiment, controlling the projection of each content onto the corresponding exhibition hall screen includes:

[0085] S141. When the screen projection content is detected to be an image, the light intensity and ambient color temperature value in the exhibition hall environment are detected.

[0086] S142. Detect the screen color temperature value of the exhibition hall screen when displaying pure white, and set the target correction color temperature value according to the ambient color temperature value;

[0087] S143. Construct the optimal function model between the target corrected color temperature value and the screen color temperature value, and use the optimal function model to calculate the correction coefficient;

[0088] S144. Adjust the saturation and brightness of the projected content according to the light intensity and correction coefficient, and adjust the color temperature value of the projected content according to the correction coefficient to generate the first projected content.

[0089] S145. Control the first projection content to be projected and displayed on the corresponding exhibition hall screen.

[0090] When the projected content is an image, the system can detect the light intensity and ambient color temperature in the exhibition hall environment. Light intensity and color temperature are important factors affecting image display quality; detecting them provides a basis for subsequent image correction. Specifically, a light sensor can be used to detect the light intensity in the exhibition hall environment, and a color temperature sensor can be used to detect the color temperature value. The color temperature sensor can measure the spectral distribution of ambient light and calculate the corresponding color temperature value.

[0091] The sensor transmits the collected light intensity and color temperature data to the system wirelessly or via wired connection. The system can receive this data in real time and store it in a local database for subsequent processing.

[0092] The system can detect the color temperature of the exhibition hall screen when displaying pure white, and set a target correction color temperature value based on the ambient color temperature to determine how to adjust the screen's color temperature to counteract the effects of ambient light. For example, optical measurement instruments can be used to measure the color temperature of the exhibition hall screen when displaying pure white, allowing for precise measurement of the screen's spectral distribution and calculation of the color temperature value.

[0093] Set the target color temperature value based on the ambient color temperature and the screen's original color temperature. For example, if the ambient color temperature is low (warm), the target color temperature can be increased appropriately to make the image look more natural under warm light.

[0094] The system constructs an optimal function model to describe the relationship between the target corrected color temperature value and the screen color temperature value, and calculates correction coefficients using this model. These correction coefficients are used to adjust the image's color temperature to achieve the target corrected color temperature value.

[0095] Specifically, a suitable function model is chosen to describe the relationship between color temperatures, such as a linear model, a polynomial model, or a nonlinear model. The parameters of the function model are then fitted using known color temperature mapping relationships. For example, the least squares method can be used to fit the model parameters, ensuring that the model accurately describes the relationship between the target corrected color temperature value and the screen color temperature value. Using the fitted function model, the correction coefficients required to derive the corrected color temperature value from the original screen color temperature value are calculated. The correction coefficients are adjustment factors used to adjust the RGB components of the image to achieve the target color temperature.

[0096] The system adjusts the image's saturation and brightness based on light intensity, and adjusts the image's color temperature value according to a correction coefficient, generating the corrected first projection content. Specifically, it adjusts the image's saturation based on light intensity. In low-light environments, it appropriately increases saturation to make the image colors more vibrant; in high-light environments, it appropriately decreases saturation to avoid oversaturation. It also adjusts the image's brightness based on light intensity. If the ambient light intensity is low, it appropriately increases the image brightness to make the image clearer; if the ambient light intensity is high, it appropriately decreases the image brightness to avoid overexposure. Finally, it adjusts the image's color temperature value based on the correction coefficient. By adjusting the image's RGB components, it brings the image's color temperature closer to the target corrected color temperature value. For example, if the target corrected color temperature value is high, it appropriately increases the blue component in the image; if the target corrected color temperature value is low, it appropriately increases the yellow component in the image.

[0097] The adjusted initial projection content is sent to the corresponding exhibition hall screen for display to ensure optimal image display. If ambient light conditions change, the system can detect and readjust the image's saturation, brightness, and color temperature in real time to ensure optimal image display under different lighting conditions.

[0098] In this embodiment, the color temperature of the ambient light in the exhibition hall affects the audience's perception of the colors of the projected image. By correcting the image color temperature to match the screen color temperature and the ambient color temperature, it is possible to effectively avoid color cast phenomena such as yellowing or bluishness caused by color temperature deviation. This ensures that the image colors can be accurately reproduced under different lighting conditions, and the color performance is more realistic and natural. This is especially important for displaying high-definition images, artworks, professional design drawings, and other content in the exhibition hall, as it can enhance the professionalism and aesthetic appeal of the images.

[0099] Secondly, if multiple screens in the exhibition hall display different projected content, it ensures color consistency across all screens, resulting in a more unified and harmonious visual effect throughout the hall. Furthermore, the lighting intensity within the exhibition hall may vary due to changes in natural light or differences in lighting in different areas. By adjusting the image's saturation and brightness according to the lighting intensity, the brightness and saturation of the image can be increased in low-light environments, making image details clearer and preventing detail loss due to overly dark images. In strong-light environments, the brightness and saturation can be appropriately reduced to prevent overexposure or overly intense colors, thus ensuring optimal visual effects under various lighting conditions, allowing viewers to clearly see image details regardless of their location within the exhibition hall.

[0100] Furthermore, appropriately adjusting saturation and brightness can not only improve image clarity but also enhance contrast, making outlines and layers more distinct. This helps in displaying complex image content (such as architectural models and data charts), making it easier for viewers to understand the information conveyed and improving information delivery efficiency. At the same time, intelligent projection control methods reduce the workload of manually adjusting image display parameters, lowering the skill requirements for exhibition staff.

[0101] Preferably, in the intelligent central control system of the exhibition hall, to accurately calculate the correction coefficients using the optimal function model, a set of randomly distributed reference correction color temperature values ​​can be initialized in the solution space of the preset initial correction coefficients. The initial correction coefficients are dynamically adjusted based on the matching degree between the display effect after correction using the reference correction color temperature values ​​and the screen color temperature values, gradually bringing the initial correction coefficients closer to the optimal solution. Finally, the system extracts the optimal correction coefficient from all converged initial correction coefficients and uses it to adjust the RGB components of the image or video, thereby achieving accurate color temperature correction and ensuring that the projected content maintains a natural and harmonious display effect under different ambient lighting conditions.

[0102] In one embodiment, controlling the projection of each content onto the corresponding exhibition hall screen includes:

[0103] When the screen-projected content is detected to be video, each video frame of the screen-projected content is divided into image blocks of a fixed size.

[0104] Find the target video frame that is most similar to each image block in the preset set of reference video frames, and calculate the motion vector of each target video frame;

[0105] The motion vector of each target video frame is scaled proportionally to form multiple intermediate frames;

[0106] Each intermediate frame is inserted between the corresponding video frames of the projected content to generate the second projected content;

[0107] Control the second projection content to be projected and displayed on the corresponding exhibition hall screen.

[0108] When the projected content is video, the system can segment each video frame into multiple fixed-size image blocks for motion estimation and compensation. These image blocks are the basic units for motion estimation. The size of the image block can be selected based on the video's resolution and motion complexity. For example, larger image blocks can be chosen for high-resolution videos to reduce computation, while smaller image blocks can be chosen for low-resolution or high-motion-complexity videos to improve accuracy.

[0109] The size of the image patch also affects the accuracy of motion estimation. Smaller image patches can capture local motion more accurately, but increase computational cost; larger image patches can reduce computational cost, but may miss some local motion details.

[0110] This embodiment uses a simple grid partitioning method to uniformly divide each video frame into multiple image blocks. For example, for a 1920×1080 resolution video frame, if a 32×32 pixel image block is selected, the video frame can be divided into 60×34 image blocks, ensuring that each image block is of the same size for uniform processing during motion estimation.

[0111] The system searches for the target video frame most similar to each image patch within a pre-defined set of reference video frames and calculates the motion vector of the target video frame relative to the current video frame. The motion vector describes the displacement of the image patch between consecutive frames. The set of reference video frames serves as a benchmark for motion estimation; the frame preceding or following the current video frame can be selected as the reference frame. For complex motion scenes, multiple reference frames can be selected to improve the accuracy of motion estimation.

[0112] Similarity measurement methods can be used to find the most similar target video frames. For example, the sum of absolute differences between the image patch and all possible positions in the reference frame can be calculated, and the position with the smallest difference can be selected as the matching position; or the normalized cross-correlation coefficient between the image patch and all possible positions in the reference frame can be calculated, and the position with the highest correlation coefficient can be selected as the matching position.

[0113] Once the most similar target video frame is found, the motion vector for each image patch is calculated. Based on the calculated motion vectors, the system scales the image patches of the target video frame proportionally, generating multiple intermediate frames. These intermediate frames fill gaps between original video frames, increasing the video's frame rate. For example, if the original video frame rate is 30fps and the target frame rate is 60fps, an intermediate frame needs to be generated between every two frames, with a scaling factor of 0.5. If the target frame rate is higher (e.g., 120fps), more intermediate frames need to be generated, with correspondingly smaller scaling factors.

[0114] The system inserts generated intermediate frames between the original video frames to form a new video sequence, i.e., the second projection content, ensuring smoother video display on the exhibition hall screen. For example, based on the timestamp of the generated intermediate frame, it is inserted between the original video frames. After insertion, the video frame rate is increased accordingly, ensuring the exhibition hall screen can support the increased frame rate and making corresponding playback adjustments. The video data after inserting the intermediate frame is re-encoded and packaged to ensure its format meets the playback requirements of the exhibition hall screen. Finally, the generated second projection content is sent to the corresponding exhibition hall screen for projection display, ensuring that viewers can watch smooth and natural video content.

[0115] This embodiment can effectively improve video smoothness and reduce motion blur by segmenting each video frame of the projected content and calculating motion vectors, generating intermediate frames that are inserted between the original video frames. This results in a more comfortable visual experience for viewers watching fast-moving scenes, without any stuttering or jerking. By scaling the intermediate frames proportionally based on the motion vectors and filling in missing pixels, flickering and jagged edges during video playback are reduced. Furthermore, since the intermediate frames are generated based on motion estimation, they can more accurately reflect the motion state of objects between frames, making the transitions between video frames more natural.

[0116] In one embodiment, controlling the projection of each content onto the corresponding exhibition hall screen includes:

[0117] Each of the projected content items is matched with the reference projected content in the content library;

[0118] When it is determined that a third type of projection content exists in the content library that matches the projection content and has superior content quality, the third type of projection content replaces the original projection content and is projected onto the corresponding exhibition hall screen; wherein, the generation method of the reference projection content includes:

[0119] The exhibit information and user tags of the exhibition hall are obtained, and the exhibit information and user tags are input into the trained projection content generation model to generate reference projection content for each exhibit under different user tags. The projection content generation model uses a convolutional neural network model as the underlying architecture.

[0120] The system matches each projected content sent by the user with reference projected content in a pre-set content library. The purpose of this matching is to find alternative content that is similar to the original but of higher quality. The content library stores optimized reference projected content, generated based on exhibit information and user tags, designed to provide a higher-quality projected experience. Exhibit information includes the exhibit's name and description. User tags reflect the user's interests, preferences, or identity information.

[0121] Feature matching algorithms can be used to determine the similarity between the original projected content and the reference projected content. For example, a convolutional neural network model can be used to extract feature vectors from an image or video and then calculate the similarity between the feature vectors.

[0122] If a reference projection content (third projection content) that matches the original projection content but is of higher quality is found in the content library, the system will replace the original projection content with this superior content and display it on the exhibition hall screen. For example, the system might assess whether the reference projection content has a higher resolution than the original; for video content, it might check whether the reference projection content has a higher frame rate; it might evaluate the encoding quality of the reference projection content, such as whether it uses a more advanced encoding format (e.g., H.265); or it might check whether the reference projection content contains more complete information, such as whether there are missing frames or parts.

[0123] The original screen-projected content is replaced with the reference screen-projected content in the screen-projection system, ensuring a seamless replacement process so that viewers will not notice any interruption.

[0124] This embodiment uses a convolutional neural network model as the underlying architecture to build a projection content generation model. This model can generate high-quality projection content based on exhibit information and user tags. Specifically, the exhibit information and user tags are preprocessed, such as image normalization and text encoding, to make them suitable for model input, ensuring that the model can generate high-quality projection content.

[0125] In this embodiment, the reference projection content is of higher quality and better adapts to the exhibition hall screen, allowing viewers to see clearer and more detailed images, thus enhancing their visual experience. Viewers of different ages and with different interests understand and appreciate the exhibits differently. The entire matching and projection process is automated, requiring no manual intervention. Exhibition hall staff do not need to manually adjust the quality and parameters of the projection content, saving time and effort and reducing operating costs.

[0126] In one embodiment, before inputting the exhibit information and user tags into the trained projection content generation model, the method further includes:

[0127] Obtain multiple samples of screen-projected content;

[0128] Each of the screen projection content samples is labeled and assigned a sample tag, which includes the exhibit information introduced by each screen projection content sample and the matching user tag;

[0129] Training data is constructed based on the multiple screen projection content samples and their corresponding sample labels;

[0130] The convolutional neural network model is iteratively trained using the training data. After the convolutional neural network model converges, the trained convolutional neural network model is used as the screen projection content generation model.

[0131] First, a large number of projection content samples are collected. These samples are used for subsequent model training to generate high-quality projection content. For example, samples can be extracted from historical projection records in the exhibition hall. These samples have been used in practice and have high reference value. Publicly available image and video datasets can also be used. These datasets are usually labeled and organized and can be directly used for model training.

[0132] Collecting a sufficient number of samples ensures adequate training and generalization ability of the model. The number of samples typically ranges from several thousand to tens of thousands, depending on the complexity of the exhibits and the diversity of the user group.

[0133] Then, the collected screen projection content samples were annotated in detail, and each sample was assigned a corresponding label. The labels included exhibit information and user tags, which were used to guide model training.

[0134] The labeled screen projection content samples and their tags are integrated into a training dataset for subsequent model training. First, noisy data, such as blurry images, unclear videos, and incorrect text descriptions, is removed from the samples; duplicate samples are deleted to avoid overfitting during model training; the sample data is normalized, such as through image normalization and text encoding, to ensure data consistency and processability; image samples are rotated, scaled, cropped, and color-adjusted to increase sample diversity, ultimately resulting in the training dataset. The cleaned and enhanced data is then converted into the format required for model training, such as converting image samples into pixel matrices and text samples into word vectors.

[0135] The convolutional neural network model is iteratively trained using a pre-constructed training dataset until it converges. For example, the training data can be divided into multiple mini-batches and fed into the model batch by batch for training. The loss value and accuracy during the training process can be monitored to identify and resolve training problems (such as overfitting and vanishing gradients) in a timely manner.

[0136] During training, the model performance is periodically evaluated using a validation set to ensure its generalization ability, and convergence conditions are set, such as the loss value no longer decreasing significantly and the accuracy stabilizing. Training is stopped early when the loss value on the validation set no longer decreases within a certain number of iterations. After the model converges, the trained convolutional neural network model is used to generate high-quality projection content.

[0137] In this embodiment, the trained model can generate high-quality projection content based on exhibit information and user tags, ensuring that images and videos are displayed more clearly and naturally on the exhibition hall screens, thus enhancing the audience's visual experience. By learning from diverse sample data, the model can adapt to different exhibits and user needs, generating personalized projection content to meet the diverse display requirements of the exhibition hall.

[0138] In one embodiment, after iteratively training the convolutional neural network model using the training data, the method further includes:

[0139] Calculate the loss value of the convolutional neural network model after training based on the cross-entropy loss function;

[0140] When the loss value is greater than a preset value, the learning rate of the convolutional neural network model is obtained, and a decay coefficient is set according to the loss value;

[0141] The learning rate of the convolutional neural network model is adjusted according to the decay coefficient, and the adjusted convolutional neural network model is retrained using the training data until the loss value is lower than the preset value, thus obtaining a trained screen projection content generation model.

[0142] During the training of a convolutional neural network model, the cross-entropy loss function is used to evaluate the model's performance. The cross-entropy loss function is a loss function particularly suitable for classification problems, measuring the difference between the model's predicted probability distribution and the true label probability distribution.

[0143] After each training iteration, the cross-entropy loss function is used to calculate the model's loss value to evaluate the training effect. If the calculated loss value is greater than a preset value, it indicates that the model's training effect has not yet reached the expected goal. At this time, it is necessary to adjust the model's learning rate to optimize the training process. The learning rate can be adjusted by setting a decay coefficient, which dynamically adjusts the learning rate based on the current loss value.

[0144] Specifically, a reasonable loss threshold is preset. When the loss value after training exceeds this threshold, a learning rate adjustment mechanism is triggered. The current learning rate is obtained from the model's training configuration, and a decay coefficient is set based on the difference between the current loss value and the preset value. For example, if the loss value is much larger than the preset value, a larger decay coefficient can be set to quickly reduce the learning rate; if the loss value is close to the preset value, a smaller decay coefficient can be set to fine-tune the learning rate. Simultaneously, a fixed decay coefficient, such as 0.1 or 0.01, is set and dynamically adjusted according to changes in the loss value.

[0145] The learning rate of the model is adjusted according to the set decay coefficient, and then the adjusted model is retrained iteratively using the same training data. This process is repeated until the loss value is lower than the preset value, and finally a trained screen projection content generation model is obtained.

[0146] This embodiment can dynamically adjust the learning rate and flexibly control its size according to the changes in the loss value. In the early stage of training, when the loss value is high, using a larger learning rate can quickly reduce the loss value and accelerate the convergence speed of the model. In the later stage of training, when the loss value is close to the preset value, using a smaller learning rate can fine-tune the model parameters, avoid the model from oscillating near the optimal solution, and ensure stable convergence of the model.

[0147] At the same time, by adjusting the learning rate, the model can better fit the training data during the training process, while avoiding overfitting. This also helps the model learn the features of the data better during training and improves the model's ability to generalize to unseen data.

[0148] Furthermore, by dynamically adjusting the learning rate using the loss value during training, the learning rate can be gradually reduced after the model has achieved a certain training effect, preventing the model from overfitting the training data, improving the robustness of the model, and ensuring that the model always adjusts in the optimal direction during training.

[0149] In one embodiment, after controlling each projected content to be displayed on the corresponding exhibition hall screen, the method further includes:

[0150] Each of the projected content items is matched with the reference projected content in the content library;

[0151] When it is determined that there is a fourth screen-casting content associated with the screen-casting content in the content library, the fourth screen-casting content is sent to the user terminal.

[0152] The system receives a confirmation instruction from the user terminal for the fourth projection content. After the projection of the original projection content corresponding to the fourth projection content is completed, the system controls the fourth projection content to be projected and displayed on the exhibition hall screen corresponding to the original projection content.

[0153] After the projected content is displayed on the exhibition hall screen, the system matches it with reference projected content in the content library to identify other content related to the current projected content. The system calculates similarity or correlation indicators by comparing the projected content with each piece of content in the reference content library, thereby finding a fourth piece of projected content that is associated with the current projected content.

[0154] Once a fourth casting content is found in the content library that is associated with the casting content, the system sends this fourth casting content to the user's device. After receiving the fourth casting content, the user can preview or review the content and decide whether to cast it.

[0155] The system receives a confirmation command from the user's terminal. After the original screen projection content finishes displaying, the system controls the fourth screen projection content to be displayed on the corresponding exhibition hall screen. This allows users to access more information related to the currently projected content, enriching their visitor experience and enabling them to gain a deeper understanding of the exhibits or themes. It also enhances user interaction with the exhibition content. This coherent content presentation helps maintain the logic and storyline of the exhibition, allowing visitors to receive information more systematically and enhancing the educational value and impact of the exhibition.

[0156] In addition, users can choose whether to view related content based on their own interests and needs, giving users a certain degree of control and making the exhibition more personalized and flexible, thus better meeting the needs of different user groups.

[0157] In one embodiment, parsing the screen mirroring request to obtain the parsed content includes:

[0158] The screen projection request is identified, and an identification result is generated. The identification result includes different content types and corresponding encoding formats in the screen projection request.

[0159] Based on the recognition results, the screen projection request is divided into multiple data blocks;

[0160] Each data block is parsed using a matching parsing algorithm, and the data blocks containing the target screen parameters are independently extracted and transformed to match the display parameters of the exhibition hall screen, thus obtaining the parsed content corresponding to each data block.

[0161] The system performs initial identification of screen mirroring requests, analyzing their content type (such as images, videos, and text) and encoding format (such as JPEG, MP4, and H.264). The identification results will serve as the basis for subsequent processing. For example, techniques such as file header recognition and MIME type detection can be used to determine the type of screen mirrored content, and the encoding format of the content can be determined by analyzing the file's encoding information, generating a structured identification result that includes information such as content type and encoding format.

[0162] Based on the recognition results, the screen projection request is divided into multiple logical data blocks. Each data block contains content of a specific type or function. Specifically, the screen projection request is divided into independent data blocks according to the content type and encoding format. For example, video content can be divided into video frame sequences, and image content can be divided into image blocks. Each data block can be appended with metadata, such as the size, location, and type of the data block, to facilitate subsequent processing.

[0163] Each data block is parsed using a parsing algorithm adapted to its content type and format. For data blocks containing target screen parameters, special processing is performed, such as extracting and converting display parameters to suit the display characteristics of the exhibition hall screen. For example, the appropriate parsing algorithm is selected based on the data block type (e.g., image, video) and encoding format (e.g., JPEG, MP4). For data blocks involving screen display parameters, necessary information (e.g., resolution, color temperature) is extracted and converted to ensure the projected content adapts to the display characteristics of the exhibition hall screen.

[0164] This embodiment can handle various types and formats of projected content, improving compatibility with different types of devices and content sources. It can flexibly adjust processing strategies according to different content types and formats, enhancing the adaptability of the projection system. Simultaneously, it performs independent data extraction and transformation processing on data blocks containing target screen parameters to adapt to the display parameters of the exhibition hall screen. Through transformation processing, it ensures that the projected content is displayed on the exhibition hall screen with optimal effect, improving the audience's visual experience. Furthermore, using the most suitable parsing algorithm can improve parsing speed, reduce processing time, and improve the overall resource utilization efficiency of the system.

[0165] Please refer to Figure 3 As shown, embodiments of the present invention also provide an intelligent central control system for exhibition halls, comprising:

[0166] The receiving module 11 is used to receive screen mirroring requests sent by the user terminal;

[0167] The parsing module 12 is used to parse the screen projection request to obtain parsed content, which includes multiple screen projection contents and content information of each screen projection content;

[0168] Evaluation module 13 is used to evaluate the display effect of each projection content on each exhibition hall screen based on the content information and exhibition hall screen information, and to calculate the adaptability of each projection content on each exhibition hall screen. The adaptability includes the evaluation value of the projection content under each evaluation index.

[0169] The projection display module 14 is used to perform a weighted summation of the evaluation values ​​of each projection content under each evaluation index to obtain the comprehensive evaluation value of each projection content on each exhibition hall screen, and to allocate exhibition hall screens for each projection content in descending order of the comprehensive evaluation values, and control each projection content to be projected and displayed on the corresponding exhibition hall screen.

[0170] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0171] In one embodiment, the present invention also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the screen projection control method described above. The storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc.

[0172] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0173] It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A screen projection control method, characterized in that, include: Receive screen mirroring requests from users; The screen mirroring request is parsed to obtain the parsed content, which includes multiple screen mirroring contents and the content information of each screen mirroring content; Based on the content information and exhibition hall screen information, the display effect of each projection content on each exhibition hall screen is evaluated, and the adaptability of each projection content on each exhibition hall screen is calculated. The adaptability includes the evaluation value of the projection content under each evaluation index. The evaluation values ​​of each projected content under each evaluation indicator are weighted and summed to obtain the comprehensive evaluation value of each projected content on each exhibition hall screen. Each projected content is assigned an exhibition hall screen in descending order of the comprehensive evaluation value, and each projected content is controlled to be projected and displayed on the corresponding exhibition hall screen. The control of displaying each projected content on the corresponding exhibition hall screen includes: Each of the projected content items is matched with the reference projected content in the content library; When it is determined that a third type of projection content exists in the content library that matches the projection content and has superior content quality, the third type of projection content replaces the original projection content and is projected onto the corresponding exhibition hall screen; wherein, the generation method of the reference projection content includes: The exhibit information and user tags of the exhibition hall are obtained, and the exhibit information and user tags are input into the trained projection content generation model to generate reference projection content for each exhibit under different user tags. The projection content generation model uses a convolutional neural network model as the underlying architecture.

2. The screen projection control method according to claim 1, characterized in that, The control of displaying each projected content on the corresponding exhibition hall screen includes: The data format of each screen-projected content is determined based on the content information; Adjust the display style of the projected content according to the data format of each projected content and the display parameters of the corresponding exhibition hall screen; The projected content is displayed on the corresponding exhibition hall screen according to the described display style.

3. The screen projection control method according to claim 1, characterized in that, The control of displaying each projected content on the corresponding exhibition hall screen includes: When the projected content is detected to be an image, the light intensity and ambient color temperature in the exhibition hall environment are detected. The screen color temperature value of the exhibition hall screen when displaying pure white is detected, and a target correction color temperature value is set according to the ambient color temperature value. Construct an optimal function model between the target corrected color temperature value and the screen color temperature value, and use the optimal function model to calculate the correction coefficients; The saturation and brightness of the projected content are adjusted according to the light intensity and correction coefficient, and the color temperature value of the projected content is adjusted according to the correction coefficient to generate the first projected content. Control the first projection content to be projected and displayed on the corresponding exhibition hall screen.

4. The screen projection control method according to claim 1, characterized in that, The control of displaying each projected content on the corresponding exhibition hall screen includes: When the screen-projected content is detected to be video, each video frame of the screen-projected content is divided into image blocks of a fixed size. Find the target video frame that is most similar to each image block in the preset set of reference video frames, and calculate the motion vector of each target video frame; The motion vector of each target video frame is scaled proportionally to form multiple intermediate frames; Each intermediate frame is inserted between the corresponding video frames of the projected content to generate the second projected content; Control the second projection content to be projected and displayed on the corresponding exhibition hall screen.

5. The screen projection control method according to claim 1, characterized in that, Before inputting the exhibit information and user tags into the trained projection content generation model, the method further includes: Obtain multiple samples of screen-projected content; Each of the screen projection content samples is labeled and assigned a sample tag, which includes the exhibit information introduced by each screen projection content sample and the matching user tag; Training data is constructed based on the multiple screen projection content samples and their corresponding sample labels; The convolutional neural network model is iteratively trained using the training data. After the convolutional neural network model converges, the trained convolutional neural network model is used as the screen projection content generation model.

6. The screen projection control method according to claim 5, characterized in that, After iteratively training the convolutional neural network model using the training data, the method further includes: Calculate the loss value of the convolutional neural network model after training based on the cross-entropy loss function; When the loss value is greater than a preset value, the learning rate of the convolutional neural network model is obtained, and a decay coefficient is set according to the loss value; The learning rate of the convolutional neural network model is adjusted according to the decay coefficient, and the adjusted convolutional neural network model is retrained using the training data until the loss value is lower than the preset value, thus obtaining a trained screen projection content generation model.

7. The screen projection control method according to claim 1, characterized in that, After controlling each projection content to be displayed on the corresponding exhibition hall screen, the method further includes: Each of the projected content items is matched with the reference projected content in the content library; When it is determined that there is a fourth screen-casting content associated with the screen-casting content in the content library, the fourth screen-casting content is sent to the user terminal. The system receives a confirmation instruction from the user terminal for the fourth projection content. After the projection of the original projection content corresponding to the fourth projection content is completed, the system controls the fourth projection content to be projected and displayed on the exhibition hall screen corresponding to the original projection content.

8. The screen projection control method according to claim 1, characterized in that, The parsing of the screen projection request yields the parsed content, including: The screen projection request is identified, and an identification result is generated. The identification result includes different content types and corresponding encoding formats in the screen projection request. Based on the recognition results, the screen projection request is divided into multiple data blocks; Each data block is parsed using a matching parsing algorithm, and the data blocks containing the target screen parameters are independently extracted and transformed to match the display parameters of the exhibition hall screen, thus obtaining the parsed content corresponding to each data block.

9. An intelligent central control system for an exhibition hall, characterized in that, include: The receiving module is used to receive screen mirroring requests sent by the user client; The parsing module is used to parse the screen projection request to obtain parsed content, which includes multiple screen projection contents and content information of each screen projection content; The evaluation module is used to evaluate the display effect of each projection content on each exhibition hall screen based on the content information and exhibition hall screen information, and to calculate the adaptability of each projection content on each exhibition hall screen. The adaptability includes the evaluation value of the projection content under each evaluation index. The projection display module is used to perform a weighted summation of the evaluation values ​​of each projection content under various evaluation indicators to obtain the comprehensive evaluation value of each projection content on each exhibition hall screen. The module then allocates exhibition hall screens to each projection content in descending order of the comprehensive evaluation values ​​and controls each projection content to be projected and displayed on the corresponding exhibition hall screen. The control of displaying each projected content on the corresponding exhibition hall screen includes: Each of the projected content items is matched with the reference projected content in the content library; When it is determined that a third type of projection content exists in the content library that matches the projection content and has superior content quality, the third type of projection content replaces the original projection content and is projected onto the corresponding exhibition hall screen; wherein, the generation method of the reference projection content includes: The exhibit information and user tags of the exhibition hall are obtained, and the exhibit information and user tags are input into the trained projection content generation model to generate reference projection content for each exhibit under different user tags. The projection content generation model uses a convolutional neural network model as the underlying architecture.

Images