Cultural heritage digital reconstruction system and method based on space narration

Abstract

The invention provides a cultural heritage digital reconstruction system and method based on space narration, and the system comprises a multi-source data collection module which is used for collecting the data of a place where cultural heritage is located; the preprocessing module is used for preprocessing the collected data; the knowledge graph construction module is used for carrying out semantic segmentation on the preprocessed data by utilizing a machine learning algorithm, carrying out classified labeling on different objects or parts, and constructing a knowledge graph of the cultural heritage according to a semantic labeling result; the spatial narrative modeling module is used for performing place analysis, event mining and path planning respectively, constructing a spatial narrative three-dimensional model of the cultural heritage by adopting polygon modeling, and adding a semantic tag for each component in the three-dimensional model; and the dynamic interaction module interacts with the spatial narrative modeling module through gestures by utilizing a gesture recognition technology, breaks through the limitation that only morphological records are concerned in traditional heritage digitization, and introduces spatial narrative into digital modeling and virtual scene design, so that the heritage becomes a lecture space.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cultural heritage data display, and in particular to a cultural heritage digital reconstruction system based on spatial narrative and a method thereof. BACKGROUND

[0002] As the living fossil of human civilization, cultural heritage carries the collective memory of different nations and different eras. However, under the multiple pressures of natural erosion, human destruction and time passing, cultural heritage is facing the risk of irreversible damage. Traditional protection methods such as physical repair and image recording can preserve the appearance but are difficult to convey the deep cultural significance. For example, the Dunhuang Mogao Grottoes murals fade due to oxidation, and the Sanxingdui bronze wares break due to rust. Traditional methods can only delay the decline but cannot restore the historical context and cultural narrative.

[0003] The existing cultural heritage profile relying too much on real-time collection weakens the value of traditional craftsmanship. Virtual repair can accurately restore the original appearance but is difficult to convey the spirit of craftsmen. Traditional display methods are single and difficult to meet the needs of the new generation for interactivity and interest. Immersive narrative is needed to enhance cultural identity.

[0004] Therefore, it is necessary to provide a new cultural heritage digital reconstruction system based on spatial narrative and a method thereof to solve the above technical problems. SUMMARY

[0005] To solve the above technical problems, the present application provides a cultural heritage digital reconstruction system based on spatial narrative and a method thereof.

[0006] The cultural heritage digital reconstruction system based on spatial narrative provided by the present application comprises a multi-source data acquisition module for acquiring data of a place where cultural heritage is located; a preprocessing module for preprocessing the acquired data; a knowledge graph construction module for performing semantic segmentation on the preprocessed data by using a machine learning algorithm, classifying and labeling different objects or parts, and constructing a knowledge graph of cultural heritage according to the semantic labeling result; a spatial narrative modeling module for performing place analysis, event mining and path planning, respectively, then constructing a spatial narrative three-dimensional model of cultural heritage by using polygon modeling according to the processed data, associating the semantic labeling result with the spatial narrative three-dimensional model, realizing digital expression of spatial form and cultural memory, and adding semantic labels to each component in the spatial narrative three-dimensional model; a dynamic interaction module for interacting with the spatial narrative modeling module by using gesture recognition technology.

[0007] Further, the multi-source data acquisition module comprises a high-precision laser scanner, a high-resolution digital camera, and a drone carrying a high-definition camera and a laser radar device. The high-precision laser scanner: multiple scanning stations are arranged around and inside the building to ensure full coverage of the site where the cultural heritage is located. The high-resolution digital camera: the cultural heritage is photographed from multiple angles, including all parts of the cultural heritage, to ensure that comprehensive two-dimensional image information is obtained, and the overlapping degree between adjacent images is controlled to be 60-80%. The drone carrying a high-definition camera and a laser radar device: according to the terrain and range of the cultural heritage and its surrounding environment, the flight route of the drone is planned, and after the surveying and mapping is completed, the high-definition images and laser radar point cloud data collected by the drone are transmitted to the preprocessing module.

[0008] Another aspect of the present application provides a method for reconstructing cultural heritage based on spatial narrative, which comprises the following steps: Step 1, data acquisition: using a high-precision laser scanner to scan the cultural heritage in all directions to obtain three-dimensional point cloud data of the target object, using professional photography equipment to photograph the cultural heritage from multiple angles to obtain high-resolution two-dimensional images, and using a drone carrying a high-definition camera and a laser radar device to conduct aerial surveying and mapping of the cultural heritage and its surrounding environment; Step 2, data processing: preprocessing the collected multi-source data; Step 3, knowledge graph construction: using a machine learning algorithm to perform semantic segmentation and classification labeling on the point cloud data, and constructing a knowledge graph of the cultural heritage according to the semantic labeling results; Step 4, spatial site analysis: analyzing the position and role of the cultural heritage and the surrounding buildings and natural landscapes; Step 5, event mining: obtaining cultural event information of oral history and traditional knowledge, and organizing and analyzing the collected cultural event information; Step 6, path planning: planning a narrative path of the cultural heritage according to the results of the site analysis and event mining; Step 7, digital modeling: creating a spatial narrative three-dimensional model of the cultural heritage according to the collected data; Step 8, spatial narrative animation interaction: based on the spatial narrative three-dimensional model, combining the results of steps 4 to 6, building a virtual scene to establish a narrative animation.

[0009] Further, the preprocessing in step 2 comprises the following steps: S1, calibrate the collected multi-source data to eliminate system errors and random errors between different devices, set multiple control points with known coordinates at the cultural heritage site, and respectively convert and register the laser scanning data and photogrammetry data to ensure accurate correspondence in space; S2, remove noise points and abnormal points in point cloud data based on distance threshold method, set distance threshold, and remove points with distance from surrounding points exceeding the threshold as noise points or abnormal points, and confirm the distance threshold by trying different distance thresholds multiple times, observing the quality change of point cloud data, and selecting the best threshold; S3, splice the point cloud data obtained by multiple scanning stations to form a complete three-dimensional point cloud model, accurately register and splice the point cloud data of different stations by using the iterative closest point algorithm, and remove excessively dense points in the point cloud while retaining the main features of the point cloud by using octree-based data simplification.

[0010] Further method, step three, semantic segmentation of point cloud data by machine learning model, training of machine learning model by using labeled sample data, application of trained model to point cloud data, classification and labeling of different objects or parts in point cloud, introduction of entity and relationship information in semantic classification and labeling result into knowledge graph by using knowledge graph construction tool, determination of entity and relationship type of knowledge graph, and formation of complete cultural heritage knowledge graph.

[0011] Further method, step four, detailed analysis of geographical position, topography, and surrounding environment of the site where the cultural heritage is located, introduction of geographical information data collected by unmanned aerial vehicle into GIS platform by using GIS platform, topographic analysis and visual display, and analysis of position and interaction of cultural heritage and surrounding buildings and natural landscape.

[0012] Further method, step five, obtaining cultural event information of oral history and traditional knowledge by consulting historical documents, archives, collecting written information related to cultural heritage, and interviewing local residents and experts and scholars, sorting and analyzing collected cultural event information, extracting key narrative elements including time, place, characters, and event process, and classifying and summarizing extracted narrative elements.

[0013] Further method, step six, planning of narrative path of cultural heritage according to results of site analysis and event mining, comprehensive consideration of spatial layout of cultural heritage and logical relationship of cultural events, hierarchical planning method, determination of main narrative main line, setting of multiple branch nodes on main line, and determination of user's tour route and narrative order in virtual scene.

[0014] Further, step seven is to construct a spatial narrative three-dimensional model of the cultural heritage according to the processed point cloud data by using a polygon modeling method, to manually create or automatically create a spatial narrative three-dimensional model of each component of the cultural heritage by using a three-dimensional modeling software as a reference and taking the point cloud data as a reference, to perform texture mapping on the spatial narrative three-dimensional model, and to give the model a real appearance and texture by taking high-resolution two-dimensional images measured by a high-resolution digital camera as a texture map.

[0015] Further, step eight is to make a narrative animation of the cultural heritage according to the narrative model and the path planning, and to make the animation by using keyframe animation and motion capture. In the keyframe animation, the key states of the cultural heritage at different narrative stages are determined, the parameters of the key frames are set in the three-dimensional modeling software, and then the intermediate frames are generated by using an animation software to form a smooth animation effect. In the motion capture, the motion capture technology is used to perform corresponding actions in a real scene, the motion data of the actors or objects are recorded in real time, and the motion data are applied to the spatial narrative three-dimensional model to generate a realistic animation, and to make a character animation or a complex object motion animation.

[0016] Compared with the related art, the spatial narrative-based cultural heritage digital reconstruction system and method provided by the application have the following beneficial effects: 1. The application analyzes the position and interaction of the cultural heritage and surrounding buildings and natural landscapes, obtains cultural event information of oral history and traditional knowledge, arranges and analyzes the collected cultural event information, plans a narrative path of the cultural heritage according to the results of the site analysis and event mining, forms a combination of spatial narrative and digital reconstruction from the site-event-path, breaks through the limitation of traditional heritage digitalization that only focuses on form recording, introduces spatial narrative into digital modeling and virtual scene design, and makes the heritage tell a space.

[0017] 2. The application uses multi-source collection, historical documents, archives, collected text information related to the cultural heritage, and interviews with local residents and experts and scholars to obtain cultural event information of oral history and traditional knowledge, and combines semantic segmentation and knowledge graph to improve the intelligent level of the model.

[0018] 3. The application explores the application of digitalization results in film and television, games, cultural and creative industries, and smart tourism by digital modeling and spatial narrative animation interaction, promotes the innovative transformation of cultural heritage in the digital economy, and forms an application mode that gives equal importance to social and economic benefits. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1A structural block diagram of a cultural heritage digital reconstruction system based on spatial narration provided by the present application is shown in the figure; Figure 2 A structural schematic diagram of a multi-source data acquisition module provided by the present application is shown in the figure; Figure 3 A flowchart of a cultural heritage digital reconstruction method based on spatial narration provided by the present application is shown in the figure; Figure 4 A flowchart of preprocessing provided by the present application is shown in the figure. DETAILED DESCRIPTION

[0020] The present application is further described below in conjunction with the accompanying drawings and embodiments.

[0021] Please refer to Figure 1 , Figure 2 , Figure 3 and Figure 4 , among which, Figure 1 A structural block diagram of a cultural heritage digital reconstruction system based on spatial narration provided by the present application is shown in the figure; Figure 2 A structural schematic diagram of a multi-source data acquisition module provided by the present application is shown in the figure; Figure 3 A flowchart of a cultural heritage digital reconstruction method based on spatial narration provided by the present application is shown in the figure; Figure 4 A flowchart of preprocessing provided by the present application is shown in the figure.

[0022] Embodiment one In the specific implementation process, as shown in the figure, Figures 1-2 The cultural heritage digital reconstruction system based on spatial narration includes a multi-source data acquisition module for acquiring data of the place where the cultural heritage is located; The multi-source data acquisition module includes a high-precision laser scanner, a high-resolution digital camera, and a drone equipped with a high-definition camera and a laser radar device. High-precision laser scanner: a high-precision laser scanner is selected, and appropriate scanning site positions and quantities are determined according to the size and complexity of the cultural heritage. For large ancient buildings, multiple scanning sites need to be set up around the building and at key positions inside the building to ensure full coverage of the target object. By setting the parameters of the laser scanner, including scanning resolution, scanning range, etc., high-resolution setting can obtain more detailed three-dimensional point cloud data, but will increase the data acquisition time and storage capacity, which needs to be balanced according to actual needs. Start the laser scanner for full-scanning, keep the device stable during scanning to avoid errors caused by shaking, and store the collected three-dimensional point cloud data in a special storage device after scanning is completed. High-resolution digital camera: Use professional photography equipment such as high-resolution digital cameras to take pictures of cultural heritage from multiple perspectives, covering all parts of the cultural heritage, including front, side, top, etc., to ensure comprehensive two-dimensional image information, control shooting distance and angle, and ensure image quality and overlap, with an overlap of 60%-80% between adjacent images to facilitate subsequent image stitching and three-dimensional reconstruction, record camera parameters such as focal length, aperture, shutter speed, and shooting position and attitude information for subsequent data processing; Unmanned aerial vehicle equipped with high-definition camera and laser radar equipment: Select unmanned aerial vehicles equipped with high-definition cameras and laser radar equipment, plan the flight route of the unmanned aerial vehicle according to the terrain and range of the cultural heritage and its surrounding environment, the flight route should cover the overall area of the cultural heritage and the surrounding environment within a certain range to obtain comprehensive spatial information, set the flight parameters of the unmanned aerial vehicle, including flight height, flight speed, and shooting interval. The flight height should be determined according to the height of the cultural heritage and the surrounding environment to ensure the acquisition of high-resolution images and point cloud data, the flight speed and shooting interval should be reasonably set to ensure the continuity and integrity of the data, and the unmanned aerial vehicle should be started to fly according to the planned route, and the status of the unmanned aerial vehicle and the data acquisition during the flight should be monitored in real time. After the surveying and mapping is completed, the high-definition images and laser radar point cloud data collected by the unmanned aerial vehicle are transmitted to the preprocessing module; Preprocessing module, preprocessing the collected data to remove noise points and abnormal points in the point cloud data to improve data quality; Knowledge graph construction module, using machine learning algorithms to perform semantic segmentation on the preprocessed data, classifying and labeling different objects or parts, and constructing a knowledge graph of the cultural heritage based on the semantic labeling results, the knowledge graph represents the relationships between entities in the cultural heritage in the form of a graph, including the connection relationships between different components in ancient buildings and the association relationships between cultural events and places; Space narrative modeling module, respectively analyzing the site, mining the event, and planning the path, then constructing a space narrative three-dimensional model of the cultural heritage using polygon modeling based on the processed data, associating the semantic labeling results with the space narrative three-dimensional model to realize the digital expression of space form and cultural memory, and adding semantic labels to each component in the space narrative three-dimensional model; Dynamic interaction module, using gesture recognition technology to interact with the space narrative modeling module, when the user clicks on a component, relevant cultural information can be displayed, and according to the narrative model and path planning, a narrative animation of the cultural heritage is produced, which vividly shows the historical events and cultural stories of the cultural heritage through keyframe animation and motion capture production.

[0023] Embodiment Two

[0024] In one implementation process, referring to Figure 3 As shown in the figure, the spatial narrative-based cultural heritage digital reconstruction method includes the following steps: Step 1, data collection: use high-precision laser scanners to scan cultural heritage from all directions to obtain three-dimensional point cloud data of the target object, use professional photography equipment to take pictures of cultural heritage from multiple angles to obtain high-resolution two-dimensional images, and use unmanned aerial vehicles equipped with high-definition cameras and laser radar equipment to conduct aerial mapping of cultural heritage and its surrounding environment; Step 2, data processing: pre-process the collected multi-source data; Step 3, knowledge graph construction: use machine learning algorithms to perform semantic segmentation and classification annotation on point cloud data, and construct a knowledge graph of cultural heritage according to the semantic annotation results; Step 4, spatial place analysis: analyze the location and role of cultural heritage and surrounding buildings and natural landscapes; Step 5, event mining: obtain cultural event information from oral history and traditional knowledge, and organize and analyze the collected cultural event information; Step 6, path planning: plan the narrative path of cultural heritage according to the results of place analysis and event mining; Step 7, digital modeling: create a spatial narrative three-dimensional model of cultural heritage based on the collected data; Step 8, spatial narrative animation interaction, based on the spatial narrative three-dimensional model, combined with the results of steps 4 to 6, build a virtual scene to establish a narrative animation.

[0025] In some embodiments, referring to Figure 4 As shown in the figure, the pre-processing in step 2 includes the following steps: S1, calibrate the collected multi-source data to eliminate system errors and random errors between different devices, set multiple control points with known coordinates at the cultural heritage site, and perform coordinate conversion and registration on laser scanning data and photogrammetry data respectively to ensure accurate correspondence of data in space; S2, remove noise points and abnormal points in point cloud data based on distance threshold method, set a distance threshold, and consider points with a distance from their surrounding points exceeding the threshold as noise points or abnormal points and remove them, the distance threshold is confirmed by trying different distance thresholds multiple times, observing the quality changes of the point cloud data, and selecting the best threshold; S3, splice the point cloud data obtained from multiple scanning stations to form a complete three-dimensional point cloud model, use the iterative closest point algorithm to accurately register and splice the point cloud data from different stations, and remove excessively dense points in the point cloud through octree-based data simplification while preserving the main features of the point cloud.

[0026] Further, step three performs semantic segmentation on the point cloud data by a machine learning model, trains the machine learning model by using the labeled sample data, applies the trained model to the point cloud data, classifies and labels different objects or parts in the point cloud, imports entity and relationship information in the semantic classification and labeling result into the knowledge graph by using a knowledge graph construction tool, determines the entity and relationship types of the knowledge graph, and forms a complete cultural heritage knowledge graph. In the knowledge graph, entities such as point cloud data of ancient buildings are labeled as different components such as walls, columns, and roofs. During the labeling process, the classification categories can be refined and adjusted according to actual conditions to ensure the accuracy and completeness of the labeling results. It should be noted that the machine learning model is trained for point cloud semantic segmentation by using labeled sample data, including the following detailed steps: Obtain point cloud data of the surface of the cultural heritage; Label the collected point cloud data and assign a semantic label to each point; Preprocess the original point cloud data, including format conversion, scaling the coordinate range of the point cloud data to a fixed size, and reducing the density of the point cloud; Select the PointNet deep learning model to directly process unordered point cloud data, extract features from the point cloud data through multiple layers of perception, and aggregate global features through a maximum pooling layer; Specify the optimizer, learning rate, and batch size parameters used during training, such as the Adam optimizer; Load the preprocessed training set data using a data loader; Train the model using the training set, update the model parameters through a backpropagation algorithm, and evaluate the model performance using a validation set during training to adjust the training strategy in a timely manner.

[0027] In some embodiments, step four analyzes the geographical location, topography, and surrounding environment of the site where the cultural heritage is located in detail, imports the geographical information data collected by the unmanned aerial vehicle into the GIS platform using the GIS platform, performs terrain analysis and visualizes the results, and analyzes the positional and functional relationships between the cultural heritage and surrounding buildings and natural landscapes.

[0028] In some embodiments, step five obtains cultural event information by consulting historical documents, archival materials, collecting written records related to cultural heritage, and interviewing local residents and experts and scholars, organizes and analyzes the collected cultural event information, extracts key narrative elements including time, location, characters, and event progression, and classifies and summarizes the extracted narrative elements.

[0029] In some embodiments, step six plans a narrative path of the cultural heritage according to the results of the site analysis and event mining, integrates the spatial layout of the cultural heritage and the logical relationship of the cultural events, adopts a hierarchical planning method, determines a main narrative main line first, and then sets multiple branch nodes on the main line to determine the tour route and narrative order of the user in the virtual scene. It should be noted that the narrative path planning focuses on user experience, enabling the user to understand the history and cultural connotation of the cultural heritage in a reasonable order. A hierarchical planning method is adopted, a main narrative main line is determined first, and then multiple branch nodes are set on the main line to provide different narrative experiences.

[0030] In some embodiments, step seven constructs a spatial narrative three-dimensional model of the cultural heritage using a polygon modeling method according to the processed point cloud data, manually creates or automatically creates a spatial narrative three-dimensional model of each component of the cultural heritage using a three-dimensional modeling software as a reference, performs texture mapping on the spatial narrative three-dimensional model, and uses high-resolution two-dimensional images measured by a high-resolution digital camera as texture maps to give the model a realistic appearance and texture.

[0031] In some embodiments, step eight produces a narrative animation of the cultural heritage according to the narrative model and path planning, produces the animation using keyframe animation and motion capture, and uses gesture recognition technology to interact with the spatial narrative modeling module. When the user clicks on a component, relevant cultural information can be displayed. Keyframe animation: determines the key states of the cultural heritage at different narrative stages, such as the movement of building components and the occurrence of cultural events, sets parameters such as position, rotation, and scaling in a three-dimensional modeling software, and then generates intermediate frames through animation software to form smooth animation effects. Motion capture: uses motion capture technology to perform corresponding actions in a real scene, records the motion data of actors or objects in real time, and applies it to the spatial narrative three-dimensional model to generate realistic animations, such as character animations or complex object motion animations.

[0032] It should be noted that gesture recognition includes the following steps: Select a depth camera or a dedicated gesture recognition sensor with gesture recognition function, which can capture the three-dimensional coordinates and motion trajectories of gestures. Set up a gesture recognition device in the display area or interaction area of the cultural heritage digitization reconstruction system to ensure that the device can capture the gesture actions of the user. Start the gesture recognition device to start collecting gesture data of the user. The user can make various gesture actions in front of the device, such as clicking, sliding, and scaling. The collected original gesture data is filtered to remove noise and outliers, improve data accuracy, and normalize gesture data to a fixed size for subsequent processing. Key features such as gesture shape, motion trajectory, and speed are extracted from preprocessed gesture data. According to the actual needs, gesture recognition based on deep learning is selected; The labeled gesture dataset is used to train the model, and the labeled dataset contains various gesture actions and their corresponding labels, so that the model can learn the mapping relationship between gestures and labels. The training set is used to train the model, and the model parameters are updated through the backpropagation algorithm until the model reaches satisfactory performance. The preprocessed gesture data is input into the trained gesture recognition model, and the model outputs the recognition result of the gesture. According to the recognition result, the system determines which gesture action the user makes, and interacts with the digital model accordingly.

[0033] The interaction logic includes the following: According to the requirements of the cultural heritage digital reconstruction system, define the mapping relationship between gesture actions and system functions, for example, click gestures can be used to select a component of cultural heritage, swipe gestures can be used to rotate or translate three-dimensional models, zoom gestures can be used to adjust the size of the model, etc. In the spatial narrative modeling module, the corresponding interaction function is called according to the gesture recognition result. When the user makes a click gesture, the system should be able to locate the component clicked by the user and display relevant cultural information such as the name, history background, and manufacturing process of the component. When the user makes a swipe or zoom gesture, the system should be able to rotate, translate, or zoom the three-dimensional model accordingly, so that the user can observe the cultural heritage from different angles.

[0034] The computing device according to the embodiments of the present application for implementing the above method includes a processor and a memory. The processor can be a multi-core processor or include multiple processors. In some embodiments, the processor can include a general-purpose main processor and one or more special-purpose coprocessors, such as a graphics processing unit (GPU), a digital signal processor (DSP), etc. In some embodiments, the processor can be implemented using custom circuits, such as an application-specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

[0035] The memory can include various types of storage units, such as a system memory, a read-only memory (ROM), and a permanent storage device. Among them, the ROM can store static data or instructions required by the processor or other modules of the computer. The permanent storage device can be a read and write storage device. The permanent storage device can be a non-volatile storage device that does not lose stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a mass storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device can be a removable storage device (such as a floppy disk, an optical drive). The system memory can be a read and write storage device or a volatile read and write storage device, such as a dynamic random access memory. The system memory can store some or all of the instructions and data required by the processor during runtime. In addition, the memory can include a combination of any computer readable storage media, including various types of semiconductor storage chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), magnetic disks and / or optical disks. In some embodiments, the memory can include a read and / or write removable storage device, such as a compact disc (CD), a read-only digital versatile disc (such as DVD-ROM, double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a min SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. The computer readable storage medium does not include carrier waves and transient electronic signals transmitted through wireless or wired transmission.

[0036] It should be understood that, unless otherwise explicitly stated herein, the execution of the above steps is not strictly limited in order, and the steps can be executed in other orders. Moreover, at least part of the steps in the processes involved in the above embodiments can include multiple steps or multiple stages, which do not necessarily be executed at the same time, but can be executed at different times, and the execution order of the steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0037] The above shows and describes the basic principles and main features of the present application and the advantages of the present application. It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, and the scope of the present application is defined by the appended claims rather than the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application, and any reference signs in the claims should not be regarded as limiting the claims involved.

[0038] Furthermore, it should be understood that although the description is made according to the embodiments, not every embodiment includes only one independent technical solution, and the description of the specification is only for the sake of clarity, and the skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be combined appropriately to form other embodiments that can be understood by the skilled in the art.

Claims

1. A digital reconstruction system for cultural heritage based on spatial narrative, characterized in that, include: A multi-source data acquisition module is used to collect data on the locations of cultural heritage sites; The preprocessing module preprocesses the collected data; The knowledge graph construction module uses machine learning algorithms to perform semantic segmentation on preprocessed data, classifies and labels different objects or parts, and constructs a knowledge graph of cultural heritage based on the semantic labeling results. The spatial narrative modeling module performs site analysis, event mining, and path planning. Then, based on the processed data, it uses polygon modeling to construct a three-dimensional spatial narrative model of cultural heritage. The semantic annotation results are associated with the three-dimensional spatial narrative model to realize the digital expression of spatial form and cultural memory. Semantic tags are added to each component in the three-dimensional spatial narrative model. The dynamic interaction module utilizes gesture recognition technology to enable interaction between the module and the spatial narrative modeling module through gestures.

2. The cultural heritage digital reconstruction system based on spatial narrative according to claim 1, characterized in that, The multi-source data acquisition module includes: a high-precision laser scanner, a high-resolution digital camera, and a drone equipped with a high-definition camera and lidar device; High-precision laser scanner: Multiple scanning stations are set up around the building and in key locations inside to ensure comprehensive coverage of the cultural heritage site; High-resolution digital camera: Captures cultural heritage from multiple perspectives, including all parts of the heritage, ensuring comprehensive two-dimensional image information is obtained; Drones equipped with high-definition cameras and lidar devices: Based on the terrain and scope of the cultural heritage and its surrounding environment, the drones' flight routes are planned. After the mapping is completed, the high-definition images and lidar point cloud data collected by the drones are transmitted to the preprocessing module.

3. A method for digital reconstruction of cultural heritage based on spatial narrative, applicable to the digital reconstruction system of cultural heritage based on spatial narrative as described in any one of claims 1-2, characterized in that, The method includes the following steps: Step 1: Data Collection: Use a high-precision laser scanner to scan the cultural heritage from all angles to obtain the three-dimensional point cloud data of the target object. Use professional photography equipment to photograph the cultural heritage from multiple perspectives to obtain high-resolution two-dimensional images. Then, use a drone equipped with a high-definition camera and lidar equipment to conduct aerial mapping of the cultural heritage and its surrounding environment. Step 2, Data Processing: Preprocess the collected multi-source data; Step 3: Knowledge Graph Construction: Use machine learning algorithms to perform semantic segmentation and classification labeling on point cloud data, and construct a knowledge graph of cultural heritage based on the semantic labeling results; Step 4: Spatial Site Analysis: Analyze the relationship between the cultural heritage site and the surrounding buildings and natural landscapes in terms of location and function; Step 5: Event Mining: Obtain cultural event information from oral history and traditional knowledge, and organize and analyze the collected cultural event information; Step Six: Route Planning: Based on the results of site analysis and event discovery, plan the narrative route for the cultural heritage; Step 7: Digital Modeling: Create a three-dimensional spatial narrative model of the cultural heritage based on the collected data; Step 8: Spatial Narrative Animation Interaction. Based on the spatial narrative 3D model and combined with the results of Steps 4 to 6, a virtual scene is built to create a narrative animation.

4. The method for digital reconstruction of cultural heritage based on spatial narrative according to claim 3, characterized in that, The preprocessing in step two includes the following steps: S1. The collected multi-source data is calibrated to eliminate systematic and random errors between different devices. Multiple control points with known coordinates are set at the cultural heritage site. Coordinate transformation and registration are performed on the laser scanning data and photogrammetric data respectively to ensure accurate spatial correspondence of the data. S2. Remove noise and outliers from point cloud data based on distance threshold method. Set a distance threshold and treat points that are more than the distance to their surrounding points as noise or outliers and remove them. Confirm the distance threshold by trying different distance thresholds multiple times, observe the quality changes of point cloud data, and select the best threshold. S3. The point cloud data obtained from multiple scanning stations are stitched together to form a complete 3D point cloud model. The iterative nearest point algorithm is used to accurately register and stitch the point cloud data from different stations. Through data simplification based on octree, overly dense points in the point cloud are removed while retaining the main features of the point cloud.

5. The method for digital reconstruction of cultural heritage based on spatial narrative according to claim 3, characterized in that, Step three involves semantic segmentation of point cloud data using a machine learning model, training the machine learning model with labeled sample data, applying the trained model to the point cloud data, classifying and labeling different objects or parts in the point cloud, and using a knowledge graph construction tool to import the entity and relationship information from the semantic classification and labeling results into the knowledge graph, determining the entity and relationship types of the knowledge graph, and forming a complete cultural heritage knowledge graph.

6. The method for digital reconstruction of cultural heritage based on spatial narrative according to claim 3, characterized in that, Step four involves a detailed analysis of the geographical location, topography, and surrounding environment of the cultural heritage site. Using a GIS platform, the geographic information data collected by the drone is imported into the GIS platform for topographic analysis and visualization, and the relationship between the location and function of the cultural heritage and the surrounding buildings and natural landscapes is analyzed.

7. The method for digital reconstruction of cultural heritage based on spatial narrative according to claim 3, characterized in that, Step five involves consulting historical documents and archives, collecting written records related to cultural heritage, and interviewing local residents and experts to obtain information on cultural events through oral history and traditional knowledge. The collected information on cultural events is then organized and analyzed to extract key narrative elements, including time, place, people, and the course of events. These extracted narrative elements are then classified and summarized.

8. The method for digital reconstruction of cultural heritage based on spatial narrative according to claim 3, characterized in that, Step six involves planning the narrative path of cultural heritage based on the results of site analysis and event mining. It integrates the spatial layout of cultural heritage with the logical relationship of cultural events, and adopts a hierarchical planning method. First, the main narrative line is determined, and then multiple branch nodes are set on the main line to determine the user's tour route and narrative order in the virtual scene.

9. The method for digital reconstruction of cultural heritage based on spatial narrative according to claim 3, characterized in that, Step seven involves constructing a spatial narrative 3D model of the cultural heritage based on the processed point cloud data using a polygon modeling method. Using 3D modeling software, the point cloud data is used as a reference to manually create or automatically generate spatial narrative 3D models of each component of the cultural heritage. Texture mapping is then performed on the spatial narrative 3D model, and high-resolution 2D images obtained by high-resolution digital cameras are used as texture maps to give the model a realistic appearance and texture.

10. The method for digital reconstruction of cultural heritage based on spatial narrative according to claim 3, characterized in that, The narrative animation in step eight is based on the narrative model and path planning to create a narrative animation of cultural heritage, using keyframe animation and motion capture to create the animation. Keyframe animation involves determining the key states of cultural heritage at different narrative stages, setting keyframe parameters in 3D modeling software, and then generating intermediate frames through animation software to create smooth animation effects. Motion capture: Using motion capture technology, corresponding actions are performed in real-world scenes, and the motion data of actors or objects is recorded in real time. This data is then applied to a 3D model of spatial narrative to generate realistic animations, creating character animations or complex object motion animations.

Images