A cultural heritage display management system and method based on cloud computing

Abstract

This invention provides a cloud computing-based cultural heritage display management system and method. The system includes a data acquisition module for acquiring multimodal data on cultural heritage; a preprocessing module for standardizing multimodal data; a feature extraction module for extracting multidimensional features of cultural heritage through algorithms; a virtual display module for generating dynamic interactive virtual scenes; a load balancing module for dynamically allocating cloud computing resources; a user analysis module for analyzing user behavior data; and a dynamic feedback module for adjusting display content in real time based on user behavior. This invention can improve user experience and system operational efficiency, while ensuring data integrity and reliability.

Description

Technical Field

[0001] The present invention relates to the field of cultural heritage protection and display, and more specifically, to a cloud computing-based cultural heritage display management system and method. Background Art

[0002] In the field of cultural heritage preservation and display, with the rapid development of information technology, digital display has gradually become an important means. Traditional display methods mainly rely on physical exhibitions and static image displays. This method is not only limited by time and space, but also fails to meet the audience's demand for a deeper understanding of cultural heritage. Although some digital display technologies have begun to be applied in recent years, most of them focus on simple image or video displays and lack interactivity and immersion. At the same time, the collection and processing of cultural heritage data faces challenges such as multi-source data fusion, noise interference, and data integrity verification. In addition, existing systems often suffer from uneven resource allocation and response delays when facing large-scale user access, affecting the user experience.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the existing technology: insufficient accuracy in data collection and processing, lack of interactivity and immersion in the display method, inflexible allocation of system resources and difficulty in coping with high concurrent user access. These problems limit the further development of the digital display of cultural heritage. Summary of the Invention

[0004] The present invention provides a cultural heritage display management system and method based on cloud computing.

[0005] In a first aspect of the present invention, a cloud computing-based cultural heritage display and management system is provided, comprising:

[0006] Data acquisition module, used to obtain multimodal data of cultural heritage;

[0007] Preprocessing module, used to standardize multimodal data;

[0008] Feature extraction module, used to extract multi-dimensional features of cultural heritage through algorithms;

[0009] Virtual display module, used to generate dynamic interactive virtual scenes;

[0010] Load balancing module, used to dynamically allocate cloud computing resources;

[0011] User analysis module, used to analyze user behavior data;

[0012] Dynamic feedback module, used to adjust display content in real time based on user behavior.

[0013] Furthermore, the data acquisition module includes:

[0014] Multi-source sensor units for collecting physical parameters of cultural heritage;

[0015] Environmental parameter unit, used to capture temperature, humidity and lighting data of the display environment;

[0016] A data verification unit, used to ensure data integrity through a verification algorithm;

[0017] The verification algorithm formula is:

[0018]

[0019] Where C is the calibration value, is the i-th type of data, is the dynamic weighting factor, is the environmental correction factor, is the timestamp parameter, is the total number of data categories.

[0020] Furthermore, the preprocessing module includes:

[0021] Noise filtering unit, which uses adaptive filtering algorithm to eliminate data noise;

[0022] A data alignment unit, used to synchronize the time axis of multimodal data;

[0023] Format standardization unit, converting data into a unified coding format;

[0024] The adaptive filtering algorithm formula is:

[0025]

[0026] Where, is the filtered data, is the original signal, is the sliding mean, is the standard deviation, is the attenuation coefficient, is the dynamic gain, is the time-dependent function.

[0027] Furthermore, the feature extraction module includes:

[0028] Spatial feature unit, extracts spatial structural features through a three-dimensional convolutional network;

[0029] Texture analysis unit, which uses a multi-scale decomposition algorithm to analyze surface texture;

[0030] Semantic association unit, constructing a correlation map between cultural heritage and historical events;

[0031] The multi-scale decomposition algorithm formula is:

[0032]

[0033] Where, To decompose the results, is the Gaussian kernel of the sth scale, For input data, is the scale weight, is the smoothing factor, is the square of the current decomposition scale.

[0034] Furthermore, the virtual display module includes:

[0035] A scene generation unit constructs a virtual space topology structure based on feature data;

[0036] Interaction logic unit, which defines the mapping relationship between user operations and scenario responses;

[0037] Rendering optimization unit, using ray tracing and detail level hybrid rendering technology;

[0038] The topology structure generation algorithm formula is:

[0039]

[0040] Where T is the optimal topology, is the feature point, are virtual coordinates, is the constraint coefficient, is the spatial curvature gradient.

[0041] Furthermore, the load balancing module includes:

[0042] Resource monitoring unit, which monitors the CPU and memory usage of each computing node in real time;

[0043] Task scheduling unit,using dynamic priority allocation algorithm;

[0044] Disaster recovery and backup unit, establishing cross-regional data mirroring;

[0045] The dynamic priority algorithm formula is:

[0046]

[0047] Where, is the CPU utilization, For the task urgency, is the memory usage, is the delay weight, For network delay.

[0048] Furthermore, the user analysis module includes:

[0049] Behavior collection unit, which records user browsing paths and duration of stay;

[0050] Preference modeling unit, constructs interest vectors with multi-dimensional features;

[0051] Prediction unit, predicts the next behavior through Markov chain;

[0052] The interest vector calculation formula is:

[0053]

[0054] Where, is the user interest vector, is the behavior type weight, is the frequency of behavior, is the time decay factor, A collection of frequency statistics for all behavior types.

[0055] Furthermore, the dynamic feedback module includes:

[0056] Content recommendation unit, matching and displaying content based on user interest vectors;

[0057] Image quality adjustment unit, dynamically adjusts rendering resolution;

[0058] Path optimization unit, reconstructs the navigation topology of the virtual scene;

[0059] The formula for dynamically adjusting rendering resolution is:

[0060]

[0061] Where, For image quality level, is the number of concurrent users, is the network bandwidth, is the terminal screen size, As the base size, is the adaptive coefficient.

[0062] Furthermore, the dynamic weighting factor The calculation methods include:

[0063] Assign initial weights based on data source reliability scores;

[0064] Update weight values ​​based on the time decay function;

[0065] Correct weight bias through feedback mechanism;

[0066] The formula is:

[0067]

[0068] Where, is the learning rate, is the actual check value, To predict the check value, is the dynamic weighting factor of the i-th type of data after the t+1th update, is the dynamic weighting factor of the i-th category data after the t-th update.

[0069] In a second aspect of the present invention, a cloud computing-based cultural heritage display and management method is provided, comprising:

[0070] S1. Collect physical parameters and environmental data of cultural heritage through multi-source sensors;

[0071] S2. Verify data integrity using a validation algorithm that includes dynamic weighting factors and environmental correction coefficients;

[0072] S3. Use adaptive filtering algorithms to reduce noise and time-align the data;

[0073] S4. Extract spatial texture features through multi-scale decomposition algorithm;

[0074] S5. Generate virtual scene structure based on topology optimization algorithm;

[0075] S6. Allocate cloud computing resources according to a dynamic priority algorithm;

[0076] S7. Construct user interest vector and predict behavior path;

[0077] S8. Adjust image quality level and content recommendation strategy in real time.

[0078] According to the above-mentioned embodiment of the present invention, there are at least the following beneficial effects: the system can ensure the integrity and reliability of cultural heritage data through the multimodal data acquisition module and data verification algorithm, and at the same time, by utilizing the adaptive filtering algorithm and data alignment unit of the pre-processing module, it can effectively eliminate data noise and achieve time axis synchronization of multi-source data, providing a high-quality data foundation for subsequent processing. In addition, the feature extraction module combines spatial features, texture analysis and semantic association units to comprehensively extract the multi-dimensional features of cultural heritage and provide rich data support for virtual display. The virtual display module adopts topology optimization algorithm and hybrid rendering technology to generate dynamic virtual scenes with strong immersion and high interactivity, thereby enhancing users' cognition and experience of cultural heritage.

[0079] The load balancing module flexibly allocates cloud computing resources through real-time monitoring of resource usage and a dynamic priority allocation algorithm, ensuring system stability and responsiveness under high-concurrency access. The cross-regional data mirroring mechanism of the disaster recovery backup unit also ensures data security and reliability. The user analysis module and dynamic feedback module record user behavior and predict their preferences in real time, dynamically adjusting display content and image quality based on user interest vectors to achieve a personalized display experience, further improving user satisfaction and the system's intelligence level. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which:

[0081] Figure 1 A schematic diagram of the structure of a cloud computing-based cultural heritage display and management system provided by one embodiment of the present invention;

[0082] Figure 2 A flowchart of a cloud computing-based cultural heritage display and management method according to an embodiment of the present invention;

[0083] Figure 3 The figure schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0084] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0085] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software.

[0086] It should be noted that any number of elements in the drawings is for illustration only and not for limitation, and any naming is only for distinction and does not have any limiting meaning.

[0087] Reference below Figure 1 , Figure 1This is a schematic diagram of the structure of a cloud computing-based cultural heritage display and management system provided by one embodiment of the present invention. Figure 1 As shown, a cloud computing-based cultural heritage display and management system 100 includes:

[0088] Data acquisition module 101, used to obtain multimodal data of cultural heritage;

[0089] A preprocessing module 102 is used to perform standardization processing on multimodal data;

[0090] Feature extraction module 103, used to extract multi-dimensional features of cultural heritage through algorithms;

[0091] A virtual display module 104 is used to generate a dynamic interactive virtual scene;

[0092] Load balancing module 105, for dynamically allocating cloud computing resources;

[0093] User analysis module 106, used to analyze user behavior data;

[0094] The dynamic feedback module 107 is used to adjust the display content in real time according to user behavior.

[0095] It should be noted that the core of this system lies in the collaborative work of multiple modules to achieve efficient display and management of cultural heritage. The data acquisition module is used to acquire multimodal data on cultural heritage, including images, videos, and 3D models. Multimodal data refers to multiple types of data collected by different sensors, such as laser scanners for 3D point cloud data, high-definition cameras for capturing images and videos, and environmental sensors for recording the temperature, humidity, and light intensity of the display environment. The preprocessing module standardizes the collected data to remove noise, align timelines, and unify the data format. The feature extraction module uses algorithms to extract multidimensional features of cultural heritage, including spatial structure, texture details, and semantic associations. The virtual display module uses the extracted features to generate dynamic, interactive virtual scenes, providing users with an immersive experience. The load balancing module dynamically allocates cloud computing resources to ensure system stability and responsiveness under high concurrent access. The user analysis module records user behavior data, builds user interest models, and predicts user behavior. The dynamic feedback module adjusts display content and image quality in real time based on user behavior to achieve personalized display effects.

[0096] Specifically, the multi-source sensor units in the data acquisition module include but are not limited to laser scanners, high-definition cameras, infrared sensors, and environmental sensors. For example, laser scanners can accurately acquire 3D point cloud data of cultural heritage sites for constructing high-precision 3D models; high-definition cameras are used to capture high-definition images and videos of cultural heritage sites, recording their detailed appearance; and environmental sensors are used to monitor the temperature, humidity, and light intensity of the display environment in real time, providing data support for subsequent display optimization. The data verification unit ensures data integrity through a specific verification algorithm. The dynamic weighting factors in the algorithm can be adjusted based on the reliability of the data source, and the environmental correction coefficient corrects the data based on environmental parameters. The noise filtering unit in the preprocessing module uses an adaptive filtering algorithm that dynamically adjusts filtering parameters using a sliding mean and standard deviation to eliminate noise in the data. The data alignment unit synchronizes the time axis of multimodal data using timestamp parameters to ensure data consistency. The format standardization unit converts data in different formats into a unified encoding format for easier processing. The spatial feature unit in the feature extraction module uses a three-dimensional convolutional network to extract the spatial structural features of cultural heritage, the texture analysis unit uses a multi-scale decomposition algorithm to analyze surface textures, and the semantic association unit constructs a correlation map between cultural heritage and historical events, providing users with richer background information.

[0097] Preferably, the multi-source sensor unit in the data acquisition module can be flexibly configured according to the type of cultural heritage and display requirements. For example, for large cultural heritage such as ancient buildings, drone cameras can be added to obtain panoramic images and videos; for small cultural relics, high-resolution microscopes can be used to obtain microscopic texture details. The verification algorithm in the data verification unit can optimize the dynamic weighting factor and environmental correction coefficient through machine learning methods to further improve the accuracy of data verification. The adaptive filtering algorithm in the preprocessing module can dynamically adjust the attenuation coefficient and dynamic gain according to the noise level of the data to achieve better noise reduction effect. The multi-scale decomposition algorithm in the feature extraction module can optimize the accuracy of texture resolution by adjusting the scale weight and smoothing factor. The scene generation unit in the virtual display module can be combined with artificial intelligence algorithms to dynamically adjust the layout and level of detail of the virtual scene according to the user's behavioral preferences. The dynamic priority algorithm in the load balancing module can dynamically adjust the resource allocation strategy according to real-time network latency and task urgency to ensure the efficient operation of the system.

[0098] In some embodiments, the data acquisition module includes:

[0099] Multi-source sensor units for collecting physical parameters of cultural heritage;

[0100] Environmental parameter unit, used to capture temperature, humidity and lighting data of the display environment;

[0101] A data verification unit, used to ensure data integrity through a verification algorithm;

[0102] The verification algorithm formula is:

[0103]

[0104] Where C is the calibration value, is the i-th type of data, is the dynamic weighting factor, is the environmental correction factor, is the timestamp parameter, is the total number of data categories.

[0105] It should be noted that the data acquisition module is one of the core components of the cultural heritage display and management system. Its main function is to obtain multimodal data on cultural heritage. Multimodal data refers to different types of data collected by multiple sensors, including physical and environmental parameters. Physical parameters refer to information such as the geometric shape, size, and material of the cultural heritage itself, while environmental parameters refer to data such as temperature, humidity, and light intensity in the cultural heritage display environment. The data verification unit ensures the integrity and accuracy of the collected data through a specific verification algorithm. This algorithm weights the data using dynamic weighting factors and environmental correction coefficients, and combines them with timestamp parameters for comprehensive verification, thus providing a reliable data foundation for subsequent data processing.

[0106] Specifically, the multi-source sensor unit in the data acquisition module may include a laser scanner, a 3D modeling device, a high-definition camera, an infrared imager, a temperature and humidity sensor, and a light sensor. The laser scanner is used to acquire 3D point cloud data of the cultural heritage, accurately reflecting its spatial structure. The high-definition camera is used to capture high-definition images of the cultural heritage, recording its appearance details. The temperature and humidity sensor and light sensor are used to monitor the temperature, humidity, and light intensity of the display environment in real time, supporting the collection of environmental parameters. The verification algorithm in the data verification unit calculates a verification value using a formula. Dynamic weighting factors can be adjusted based on the reliability of the data source. For example, a higher weight can be assigned to high-precision laser scanner data. The environmental correction factor is adjusted based on changes in environmental parameters to compensate for the impact of environmental factors on data collection. The timestamp parameter is used to mark the chronological order of data collection, ensuring temporal consistency. For example, if the timestamps of the acquired image data and the environmental data do not match, the environmental correction factor can be adjusted to correct the data.

[0107] Preferably, the data acquisition module can be flexibly configured according to the specific type and display requirements of the cultural heritage. For large cultural heritage, such as ancient buildings or ruins, high-definition cameras and laser scanners carried by drones can be added to obtain more comprehensive three-dimensional data and panoramic images; for small cultural relics, high-resolution microscopes and photometric stereo vision equipment can be used to obtain micro-texture and material information. The verification algorithm in the data verification unit can optimize the dynamic weighting factor and environmental correction coefficient through machine learning methods. For example, a neural network trained based on historical data can predict the reliability of data from different sensors and dynamically adjust the weights, thereby further improving the accuracy of data verification. In addition, timestamp parameters can be recorded in combination with blockchain technology to ensure the non-tamperability and traceability of data, providing more reliable protection for the digital protection of cultural heritage.

[0108] In some embodiments, the pre-processing module includes:

[0109] Noise filtering unit, which uses adaptive filtering algorithm to eliminate data noise;

[0110] A data alignment unit, used to synchronize the time axis of multimodal data;

[0111] Format standardization unit, converting data into a unified coding format;

[0112] The adaptive filtering algorithm formula is:

[0113]

[0114] Where, is the filtered data, is the original signal, is the sliding mean, is the standard deviation, is the attenuation coefficient, is the dynamic gain, is the time-dependent function.

[0115] It should be noted that the preprocessing module plays a vital role in the cultural heritage display and management system. Its main function is to standardize the collected multimodal data to improve data availability and consistency. The preprocessing module includes a noise filtering unit, a data alignment unit, and a format standardization unit. These units use specific algorithms and techniques to ensure that the data meets high-quality standards before entering the subsequent processing flow. Among them, the noise filtering unit uses an adaptive filtering algorithm that can dynamically adjust the filtering parameters according to the characteristics of the data, thereby effectively eliminating noise; the data alignment unit is responsible for synchronizing data from different sources on the timeline to ensure data consistency; and the format standardization unit converts the data into a unified encoding format for subsequent processing and storage.

[0116] Specifically, the adaptive filtering algorithm in the noise filtering unit is based on a comprehensive process using a sliding mean, standard deviation, attenuation coefficient, dynamic gain, and a time-correlation function. The sliding mean smooths short-term fluctuations in the data, the standard deviation measures the degree of data dispersion, and the attenuation coefficient and dynamic gain adjust the filter's strength and response speed. The time-correlation function dynamically adjusts the filtering effect based on the temporal characteristics of the data, thereby better adapting to different types of noise. For example, when processing image data of cultural heritage, if Gaussian noise is present, the filtering effect can be optimized by adjusting the attenuation coefficient and dynamic gain. The data alignment unit synchronizes the time axis of multimodal data using timestamp parameters, ensuring temporal consistency of data collected by different sensors. The format standardization unit converts data into a unified encoding format, such as JPEG or PNG for image data and PLY or OBJ for 3D point cloud data, to facilitate processing and storage in subsequent modules.

[0117] Preferably, the parameters in the adaptive filtering algorithm can be optimized according to the specific application scenario. For example, when processing video data of cultural heritage, the attenuation coefficient can be set to 0.1 and the dynamic gain can be set to 1.5 to better adapt to the dynamic characteristics of the video data. For high-resolution image data, the window size of the sliding mean can be appropriately increased to improve the effect of noise filtering. The data alignment unit can introduce a time correction algorithm to further optimize the accuracy of time alignment through interpolation or compensation methods. For example, when the sampling frequencies of different sensors are different, the low-frequency data can be time-compensated by interpolation methods to align it with the high-frequency data. In addition, the format standardization unit can support the conversion of multiple encoding formats to meet the needs of different users. For example, for data that needs to be transmitted over the network, a format with a higher compression rate can be selected; for data that needs to be stored locally, a lossless compression format can be selected.

[0118] In some embodiments, the feature extraction module includes:

[0119] Spatial feature unit, extracts spatial structural features through a three-dimensional convolutional network;

[0120] Texture analysis unit, which uses a multi-scale decomposition algorithm to analyze surface texture;

[0121] Semantic association unit, constructing a correlation map between cultural heritage and historical events;

[0122] The multi-scale decomposition algorithm formula is:

[0123]

[0124] Where, To decompose the results, is the Gaussian kernel of the sth scale, For input data, is the scale weight, is the smoothing factor, is the square of the current decomposition scale.

[0125] It should be noted that the feature extraction module is a key link in the cultural heritage display and management system. Its main function is to extract the multi-dimensional features of cultural heritage from multimodal data through algorithms. These features include spatial structure features, texture features, and semantic association features, which can provide rich information support for subsequent virtual display and user interaction. The spatial feature unit uses a three-dimensional convolutional network to extract the spatial structure features of cultural heritage, and can identify the geometric shape and spatial layout of cultural heritage; the texture analysis unit uses a multi-scale decomposition algorithm to analyze surface textures, and can capture the detailed texture information of cultural heritage; and the semantic association unit constructs a correlation map between cultural heritage and historical events, providing users with a deeper cultural background and historical information.

[0126] Specifically, the three-dimensional convolutional network in the spatial feature unit is a deep learning algorithm that can process three-dimensional data and extract its spatial structural features. The network extracts features from the three-dimensional model of cultural heritage through a combination of convolutional layers, pooling layers, and fully connected layers, thereby identifying the geometric shape and spatial layout of the cultural heritage. The multi-scale decomposition algorithm in the texture analysis unit is an algorithm based on wavelet transform. It decomposes the input data through Gaussian kernels of different scales and can capture detailed information about the surface texture of cultural heritage. The semantic association unit associates cultural heritage with related historical events, cultural background and other information by constructing a knowledge graph, providing users with a richer cultural experience. For example, for an ancient cultural relic, the semantic association unit can be used to associate it with the dynasty, historical figures or major events to which it belongs, thereby enhancing the user's understanding of the cultural relic.

[0127] Preferably, the structure of the three-dimensional convolutional network can be optimized according to the characteristics of the cultural heritage data. For example, the number of convolutional layers can be increased to extract more complex geometric features, or the size of the convolution kernel can be adjusted to adapt to cultural heritage of different scales. The multi-scale decomposition algorithm in the texture analysis unit can optimize the effect of texture extraction by adjusting the scale weight and smoothing factor. For example, for high-resolution image data, the scale weight can be increased to extract finer texture features; for low-resolution data, the scale weight can be appropriately reduced to avoid overfitting. In addition, the semantic association unit can introduce natural language processing technology to further enrich the historical background information of cultural heritage through text mining and semantic analysis. For example, historical documents and archaeological reports can be combined to generate a detailed semantic description for cultural heritage to enhance the user's cultural experience.

[0128] In some embodiments, the virtual display module includes:

[0129] A scene generation unit constructs a virtual space topology structure based on feature data;

[0130] Interaction logic unit, which defines the mapping relationship between user operations and scenario responses;

[0131] Rendering optimization unit, using ray tracing and detail level hybrid rendering technology;

[0132] The topology structure generation algorithm formula is:

[0133]

[0134] Where T is the optimal topology, is the feature point, are virtual coordinates, is the constraint coefficient, is the spatial curvature gradient.

[0135] It should be noted that the virtual display module is the core component of the cultural heritage display and management system for user interaction and immersive experiences. Its function is to generate dynamic, interactive virtual scenes based on extracted feature data. This module uses a scene generation unit to construct the topological structure of the virtual space. The interaction logic unit defines the mapping between user operations and scene responses. The rendering optimization unit uses advanced rendering technology to enhance the visual effects of the scene. The topology generation algorithm optimizes the relationship between feature points and virtual coordinates to ensure that the spatial layout of the virtual scene is reasonable and consistent with the actual structure of the cultural heritage, thereby providing users with a highly realistic display effect.

[0136] Specifically, the scene generation unit constructs the topological structure of the virtual space based on the characteristic data of the cultural heritage. The topological structure generation algorithm generates the optimal virtual space layout by optimizing the relationship between feature points and virtual coordinates, combining the constraint coefficient and the spatial curvature gradient. For example, for the virtual display of ancient buildings, the algorithm will generate a virtual topological structure that is highly consistent with the actual building based on the three-dimensional feature points and spatial structure of the building. The interactive logic unit defines the mapping relationship between user operations and scene responses. For example, users can move, zoom and rotate in the virtual scene through gestures or mouse operations, and the system responds to user operations in real time according to the preset interactive logic. The rendering optimization unit uses ray tracing and detail level hybrid rendering technology, combined with the texture and material information of cultural heritage, to generate high-quality visual effects. For example, for cultural relics with complex textures, the rendering optimization unit can dynamically adjust the rendering accuracy through detail level technology to ensure the best visual effects at different viewing distances.

[0137] Preferably, the constraint coefficient in the topology generation algorithm can be adjusted according to the type of cultural heritage and the display requirements. For example, for cultural relics that need to be displayed with high precision, the constraint coefficient can be appropriately increased to ensure the accuracy of the topology; for large cultural heritage such as ancient buildings, the constraint coefficient can be appropriately reduced to optimize computing efficiency. The interactive logic unit can introduce artificial intelligence algorithms to dynamically adjust the interactive response strategy according to the user's operating habits. For example, the user's next operation can be predicted through a machine learning algorithm, and the scene loading and response mechanism can be optimized in advance to improve the user experience. The rendering optimization unit can dynamically adjust the rendering strategy in combination with real-time hardware performance monitoring. For example, full ray tracing technology is enabled on high-performance devices, and switched to simplified rendering mode on low-performance devices to ensure system compatibility and smoothness.

[0138] In some embodiments, the load balancing module includes:

[0139] Resource monitoring unit, which monitors the CPU and memory usage of each computing node in real time;

[0140] Task scheduling unit,using dynamic priority allocation algorithm;

[0141] Disaster recovery and backup unit, establishing cross-regional data mirroring;

[0142] The dynamic priority algorithm formula is:

[0143]

[0144] Where, is the CPU utilization, For the task urgency, is the memory usage, is the delay weight, For network delay.

[0145] It should be noted that the load balancing module is a key component of the cultural heritage display and management system, used to optimize cloud computing resource allocation and ensure stable system operation. Its function is to monitor the resource usage of each computing node in real time, use a dynamic priority allocation algorithm to rationally schedule tasks, and establish cross-regional data mirroring for disaster recovery and backup. The resource monitoring unit is responsible for obtaining key metrics such as the CPU and memory usage of computing nodes in real time. The task scheduling unit dynamically assigns task priorities based on these metrics, ensuring efficient system operation under high-concurrency access. The disaster recovery and backup unit uses cross-regional data mirroring technology to ensure data security and reliability, preventing data loss or service interruptions due to single points of failure.

[0146] Specifically, the resource monitoring unit collects key metrics such as CPU utilization and memory usage in real time through monitoring programs deployed on each compute node. These metrics are reported to the task scheduling unit at a fixed frequency (e.g., once per second). The task scheduling unit calculates the priority of each task based on a dynamic priority algorithm, combining factors such as CPU utilization, task urgency, memory usage, and network latency. For example, when a compute node experiences high CPU utilization, the algorithm will reassign new tasks to other nodes with lower loads, thereby achieving balanced resource utilization. The disaster recovery unit ensures redundant data storage by deploying data mirrors in different geographic regions. If a server in one region fails, the system can quickly failover to a data mirror in another region, ensuring service continuity. The parameters in the dynamic priority algorithm can be adjusted based on the actual application scenario. For example, latency weighting can be optimized based on network stability, and task urgency can be categorized by the type of user request (e.g., real-time interaction or offline browsing).

[0147] Preferably, the resource monitoring unit can introduce machine learning technology to predict resource usage trends based on historical data and adjust resource allocation strategies in advance. For example, based on a time series analysis model, resource demand during peak hours can be predicted and computing resources can be reserved in advance. The task scheduling unit can combine containerization technologies (such as Docker and Kubernetes) to achieve rapid deployment and elastic expansion of tasks. For example, when a node is detected to be overloaded, the system can automatically start a new container instance to share the task. The disaster recovery backup unit can use distributed storage technology (such as Ceph or GlusterFS) to further improve data reliability and read and write performance. In addition, the dynamic priority algorithm can introduce user behavior analysis to dynamically adjust task priorities based on user activity and behavior patterns. For example, higher priority can be assigned to requests from high-frequency users or important customers to improve user experience.

[0148] In some embodiments, the user analysis module includes:

[0149] Behavior collection unit, which records user browsing paths and duration of stay;

[0150] Preference modeling unit, constructs interest vectors with multi-dimensional features;

[0151] Prediction unit, predicts the next behavior through Markov chain;

[0152] The interest vector calculation formula is:

[0153]

[0154] Where, is the user interest vector, is the behavior type weight, is the frequency of behavior, is the time decay factor, A collection of frequency statistics for all behavior types.

[0155] It should be noted that the user analysis module is a crucial component of the cultural heritage display and management system, used to enhance user experience and achieve personalized presentations. This module uses a behavior collection unit to record user behavior data such as browsing paths and dwell time. It utilizes a preference modeling unit to construct a multidimensional interest vector for users. Furthermore, a prediction unit uses a Markov chain to predict the user's next action. These functions enable the system to dynamically adjust display content and interaction strategies based on user interests and behaviors, thereby providing more personalized services. The calculation of the interest vector incorporates behavior type weights, behavior frequency, and a time decay factor to accurately reflect user preferences.

[0156] Specifically, the behavior collection unit deploys monitoring programs on user terminals and servers to record data such as users' browsing paths, dwell time, and click behavior in virtual display scenes. This data is timestamped and stored in the system database. The preference modeling unit constructs a multidimensional interest vector for users based on the collected behavior data. For example, behavior type weights can be assigned based on the user's dwell time and interaction frequency on different categories of content (such as cultural relics, historical background, virtual scenes, etc.). Behavior frequency counts the number of times a user visits specific content, and the time decay factor is used to reduce the impact of a user's past behavior on their current interests. The prediction unit uses a Markov chain model to predict a user's likely next behavior based on their historical behavior sequence. For example, if a user frequently clicks on relevant historical background information while browsing cultural relics, the system can predict that they are likely to continue exploring similar content.

[0157] Preferably, the behavior collection unit can be combined with eye tracking technology to further obtain the user's visual focus data on the displayed content, so as to capture the user's interest more accurately. The preference modeling unit can introduce deep learning algorithms, such as neural networks, to perform more complex feature extraction and modeling on user behavior data to adapt to the behavior patterns of different users. For example, through user profiling technology, users can be divided into different interest groups, and personalized display strategies can be customized for each group. The prediction unit can be combined with reinforcement learning algorithms to dynamically adjust the prediction model according to real-time user feedback to improve the accuracy and adaptability of the prediction. In addition, the calculation of the interest vector can introduce more dimensional features, such as the user's social network behavior, geographic location information, etc., to more comprehensively reflect the user's preferences.

[0158] In some embodiments, the dynamic feedback module includes:

[0159] Content recommendation unit, matching and displaying content based on user interest vectors;

[0160] Image quality adjustment unit, dynamically adjusts rendering resolution;

[0161] Path optimization unit, reconstructs the navigation topology of the virtual scene;

[0162] The formula for dynamically adjusting rendering resolution is:

[0163]

[0164] Where, For image quality level, is the number of concurrent users, is the network bandwidth, is the terminal screen size, As the base size, is the adaptive coefficient.

[0165] It should be noted that the dynamic feedback module is a key component of the cultural heritage display and management system for enhancing user experience and achieving personalized presentations. This module dynamically adjusts display content, image quality levels, and virtual scene navigation paths based on user behavior and preferences through a content recommendation unit, image quality adjustment unit, and path optimization unit. The content recommendation unit matches display content based on user interest vectors, the image quality adjustment unit dynamically adjusts rendering resolution based on network status and user device, and the path optimization unit reconstructs the virtual scene's navigation topology to enhance the user interaction experience. These functions enable the system to respond to user needs in real time, providing a more immersive and personalized presentation experience.

[0166] Specifically, the content recommendation unit matches the display content based on the user interest vector, where the user interest vector is a multidimensional feature vector generated by the user analysis module, which reflects the user's interest in different types of cultural heritage. The image quality adjustment unit dynamically adjusts the rendering resolution according to the number of concurrent users, network bandwidth and user terminal screen size to ensure that the best visual effects can be provided in different network environments. For example, when the network bandwidth is low, the system will automatically reduce the image quality level to reduce lag; when the user uses a high-resolution screen, the system will increase the image quality level to enhance the visual experience. The path optimization unit reconstructs the navigation topology of the virtual scene by analyzing the user's browsing path and stop points, providing users with a more intuitive and convenient interaction path. For example, if a user stays in a specific area for a long time, the system can automatically adjust the navigation path to guide the user to explore related content.

[0167] Preferably, the content recommendation unit can introduce a collaborative filtering algorithm, combined with the behavioral data of other users, to provide more accurate recommended content for the current user. For example, if the system detects that most users will check the historical background of a certain cultural relic after browsing it, the system can automatically provide a similar recommended path for the new user. The image quality adjustment unit can be combined with real-time network monitoring technology to dynamically adjust the adaptive coefficient of the image quality level. For example, by monitoring network latency and packet loss rate, the system can automatically switch to low-quality mode when the network is unstable, and gradually improve the image quality after the network is restored. The path optimization unit can introduce artificial intelligence algorithms, such as reinforcement learning, to dynamically adjust the navigation path based on real-time user feedback to further enhance the user experience. In addition, the image quality adjustment strategy can introduce a user feedback mechanism to allow users to manually adjust their image quality preferences and incorporate these preferences into the system's automatic adjustment strategy.

[0168] In some embodiments, the dynamic weighting factor The calculation methods include:

[0169] Assign initial weights based on data source reliability scores;

[0170] Update weight values ​​based on the time decay function;

[0171] Correct weight bias through feedback mechanism;

[0172] The formula is:

[0173]

[0174] Where, is the learning rate, is the actual check value, To predict the check value, is the dynamic weighting factor of the i-th type of data after the t+1th update, is the dynamic weighting factor of the i-th category data after the t-th update.

[0175] It should be noted that the calculation method of the dynamic weighting factor is a key step in the data verification unit, used to ensure data integrity and reliability. The calculation of the dynamic weighting factor combines the data source reliability score, the time decay function, and a feedback mechanism, enabling dynamic adjustment of the weight value based on the real-time status of the data. The data source reliability score assesses the quality of data collected by different sensors, the time decay function is used to reduce the impact of old data on the current weight, and the feedback mechanism corrects weight deviations by comparing the actual verification value with the predicted verification value. This method can effectively adapt to dynamic changes in the data collection process and improve the accuracy and adaptability of data verification.

[0176] Specifically, the initial weight assignment of the dynamic weighting factor is based on the reliability score of the data source. For example, for high-precision laser scanner data, the initial weight can be set to a higher value (such as 0.8), while for data collected by environmental sensors, the initial weight can be set to a lower value (such as 0.5). The time decay function is used to adjust the weight value. As the data collection time goes by, the weight will gradually decay to reduce the impact of old data on the current verification. For example, the time decay function can be set to an exponential decay form, such as ,in is the attenuation coefficient ( ), is the time step. The feedback mechanism corrects the weight deviation by comparing the difference between the actual check value and the predicted check value. For example, if the difference between the actual check value and the predicted check value is large, it means that the current weight setting is unreasonable and needs to be adjusted by the learning rate. Adjust the weights, the formula is

[0177]

[0178] in, is the current weight, is the predicted checksum value, is the actual checksum value.

[0179] Preferably, the calculation of dynamic weighting factors can introduce machine learning algorithms, such as linear regression or neural networks, to more accurately predict weight changes. For example, by training the model with historical data, the weight values ​​can be dynamically adjusted to adapt to the reliability changes of different data sources. The time decay function can be adjusted according to the data type and application scenario. For example, for data with higher real-time requirements, a faster decay rate can be used; for data with higher stability, a slower decay rate can be used. In addition, the learning rate in the feedback mechanism It can be dynamically adjusted based on the system's fault tolerance and the accuracy requirements of data verification. For example, in scenarios where data verification accuracy is high, the learning rate can be set to a higher value to quickly correct weight deviations; in scenarios where fault tolerance is high, the learning rate can be set to a lower value to avoid over-adjustment.

[0180] The above-described embodiments of the present invention have the following beneficial effects: Through the multi-source sensor unit and environmental parameter unit of the data acquisition module, the system can acquire comprehensive multimodal data on cultural heritage, while utilizing the data verification unit to ensure data integrity and accuracy. The noise filtering unit, data alignment unit, and format standardization unit of the preprocessing module effectively eliminate data noise, synchronize timelines, and unify data formats, thereby providing a high-quality data foundation for subsequent processing. The spatial feature unit, texture analysis unit, and semantic association unit of the feature extraction module accurately extract the multidimensional features of cultural heritage, providing a comprehensive analysis of spatial structure, texture details, and historical associations. The scene generation unit, interaction logic unit, and rendering optimization unit of the virtual display module can construct immersive, dynamic, interactive virtual scenes, combining ray tracing with level-of-detail hybrid rendering technology to enhance the user experience. The resource monitoring unit, task scheduling unit, and disaster recovery unit of the load balancing module monitor resource usage in real time, dynamically allocate cloud computing resources, and ensure data security through cross-regional data mirroring, ensuring system stability and reliability under high concurrent access. The behavior acquisition unit, preference modeling unit, and prediction unit of the user analysis module record user behavior in real time, construct user interest vectors, and predict future behavior, providing data support for personalized display. The content recommendation unit, image quality adjustment unit and path optimization unit of the dynamic feedback module can dynamically adjust the display content, image quality level and navigation topology according to user interests, further improving user satisfaction and the intelligence level of the system.

[0181] Furthermore, the dynamic weighting factor calculation method assigns initial weights based on the data source reliability score and updates the weights in real time based on a time decay function and feedback mechanism, further improving the accuracy and adaptability of data verification. Overall, this system enables efficient collection, precise processing, immersive display, and intelligent management of cultural heritage, while ensuring data integrity and system stability, providing comprehensive technical support for the digital preservation and display of cultural heritage.

[0182] like Figure 2 As shown, some embodiments provide a cloud computing-based cultural heritage display management method 200, the method 200 including:

[0183] S1. Collect physical parameters and environmental data of cultural heritage through multi-source sensors;

[0184] S2. Verify data integrity using a validation algorithm that includes dynamic weighting factors and environmental correction coefficients;

[0185] S3. Use adaptive filtering algorithms to reduce noise and time-align the data;

[0186] S4. Extract spatial texture features through multi-scale decomposition algorithm;

[0187] S5. Generate virtual scene structure based on topology optimization algorithm;

[0188] S6. Allocate cloud computing resources according to a dynamic priority algorithm;

[0189] S7. Construct user interest vector and predict behavior path;

[0190] S8. Adjust image quality level and content recommendation strategy in real time.

[0191] It is understandable that the steps described in the cloud computing-based cultural heritage display management method 200 are similar to those in the reference Figure 1 Therefore, the modules, features, and beneficial effects described above for the cloud-based cultural heritage display management system are also applicable to the cloud-based cultural heritage display management method 200 and the operations contained therein, and will not be repeated here.

[0192] Reference below Figure 3 , which shows a schematic structural diagram of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0193] like Figure 3 As shown, electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 302 or programs loaded from a storage device 308 into a random access memory (RAM) 303. RAM 303 also stores various programs and data required for the operation of electronic device 300. Processing device 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to bus 304.

[0194] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as needed.

[0195] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or a terminal device such as a computer, server, mobile phone, or tablet.

[0196] The above descriptions merely illustrate some preferred embodiments of the present invention and the underlying technical principles. Those skilled in the art should understand that the scope of the invention encompassed by the embodiments of the present invention is not limited to technical solutions formed by specific combinations of the aforementioned technical features. It also encompasses other technical solutions formed by any combination of the aforementioned technical features or their equivalents, without departing from the aforementioned inventive concept. For example, a technical solution formed by replacing the aforementioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A cultural heritage display and management system based on cloud computing, characterized in that: include: Data acquisition module, used to obtain multimodal data of cultural heritage; Preprocessing module, used to standardize multimodal data; Feature extraction module, used to extract multi-dimensional features of cultural heritage through algorithms; Virtual display module, used to generate dynamic interactive virtual scenes; Load balancing module, used to dynamically allocate cloud computing resources; User analysis module, used to analyze user behavior data; Dynamic feedback module, used to adjust the display content in real time according to user behavior; The feature extraction module includes: Spatial feature unit, extracts spatial structural features through a three-dimensional convolutional network; Texture analysis unit, which uses a multi-scale decomposition algorithm to analyze surface texture; Semantic association unit, constructing a correlation map between cultural heritage and historical events; The multi-scale decomposition algorithm formula is: Where M out is the decomposition result, G s is the Gaussian kernel of the sth scale, I is the input data, λ s is the scale weight, θ is the smoothing factor, s 2 is the square of the current decomposition scale.

2. The system according to claim 1, wherein: The data acquisition module includes: Multi-source sensor units for collecting physical parameters of cultural heritage; Environmental parameter unit, used to capture temperature, humidity and lighting data of the display environment; A data verification unit, used to ensure data integrity through a verification algorithm; The verification algorithm formula is: Where C is the calibration value, D i is the i-th type of data, W i is the dynamic weighting factor, α is the environmental correction coefficient, E t is the timestamp parameter, and n is the total number of data categories.

3. The system according to claim 1, wherein: The pre-processing module comprises: Noise filtering unit, which uses adaptive filtering algorithm to eliminate data noise; A data alignment unit, used to synchronize the time axis of multimodal data; Format standardization unit, converting data into a unified coding format; The adaptive filtering algorithm formula is: F k =β·(S raw μ) / σ+γ·δ(t); Where, F k is the filtered data, S raw is the original signal, μ is the sliding mean, σ is the standard deviation, β is the attenuation coefficient, γ is the dynamic gain, and δ(t) is the time correlation function.

4. The system according to claim 1, wherein: The virtual display module includes: A scene generation unit constructs a virtual space topology structure based on feature data; Interaction logic unit, which defines the mapping relationship between user operations and scenario responses; Rendering optimization unit, using ray tracing and detail level hybrid rendering technology; The formula for constructing the virtual space topology structure based on feature data is: Where T is the optimal topology, P i is the feature point, Q j is the virtual coordinate, η is the constraint coefficient, is the spatial curvature gradient.

5. The system according to claim 1, wherein: The load balancing module includes: Resource monitoring unit, which monitors the CPU and memory usage of each computing node in real time; Task scheduling unit,using dynamic priority allocation algorithm; Disaster recovery and backup unit, establishing cross-regional data mirroring; The dynamic priority allocation algorithm formula is: Where U c is the CPU utilization, τ is the task urgency, M used is the memory usage, ω is the delay weight, D latency For network delay.

6. The system according to claim 1, wherein: The user analysis module includes: Behavior collection unit, which records user browsing paths and duration of stay; Preference modeling unit, constructs interest vectors with multi-dimensional features; Prediction unit, predicts the next behavior through Markov chain; The interest vector calculation formula is: Where V user is the user interest vector, φ k is the behavior type weight, A k is the frequency of behavior, B k is the time decay factor, A is the set of frequency statistics of all behavior types; m is the total number of behavior types performed by the user.

7. The system according to claim 1, wherein: The dynamic feedback module includes: Content recommendation unit, matching and displaying content based on user interest vectors; Image quality adjustment unit, dynamically adjusts rendering resolution; Path optimization unit, reconstructs the navigation topology of the virtual scene; The formula for dynamically adjusting rendering resolution is: Where Q level is the image quality level, N user is the number of concurrent users, R bandwidth is the network bandwidth, S device is the terminal screen size, S base is the base size, and ∈ is the adaptive coefficient.

8. The system according to claim 2, wherein: The dynamic weighting factor W i The calculation methods include: Assign initial weights based on data source reliability scores; Update weight values ​​based on the time decay function; Correct weight bias through feedback mechanism; The formula is: Where ζ is the learning rate, E t is the actual check value, To predict the check value, is the dynamic weighting factor of the i-th type of data after the t+1th update, is the dynamic weighting factor of the i-th category data after the t-th update.

9. A cultural heritage display and management method based on cloud computing, characterized in that: The following steps are involved: S1. Collect physical parameters and environmental data of cultural heritage through multi-source sensors; S2. Verify data integrity using a validation algorithm that includes dynamic weighting factors and environmental correction coefficients; S3. Use adaptive filtering algorithms to reduce noise and time-align the data; S4. Extract spatial texture features through multi-scale decomposition algorithm; S5. Generate virtual scene structure based on topology optimization algorithm; S6. Allocate cloud computing resources according to a dynamic priority algorithm; S7. Construct user interest vector and predict behavior path; S8. Real-time adjustment of image quality level and content recommendation strategy; The extracting of spatial texture features by a multi-scale decomposition algorithm includes: Extract spatial structural features through 3D convolutional networks; Use multi-scale decomposition algorithm to analyze surface texture; Construct a map of the relationship between cultural heritage and historical events; The multi-scale decomposition algorithm formula is: Where M out is the decomposition result, G s is the Gaussian kernel of the sth scale, I is the input data, λ s is the scale weight, θ is the smoothing factor, s 2 is the square of the current decomposition scale.

Images