An immersive safety accident reproduction system and method based on digital twins

Abstract

The present invention relates to the technical field of accident reconstruction, and specifically discloses an immersive safety accident reconstruction system and method based on digital twins. The system includes: a scene rendering end, which builds a refined original scene rendering model based on the facility texture, three-dimensional remote sensing terrain data and the original model of preset scene facilities in the accident scene video; a dynamic fusion rendering end, which dynamically renders the multi-attribute, motion tracking and accident development characterization data of the characters at the accident scene by fusing them with the refined original scene rendering model to generate a safety accident reconstruction dynamic model; an immersive display end, which divides the safety accident reconstruction dynamic model based on the key moments of the accident process to obtain corresponding local models, and then displays single or multiple local models to the user through virtual reality or augmented reality equipment; thereby improving the accuracy, authenticity and user experience of safety accident reconstruction, helping to improve the level of accident safety management, and enhancing the user's immersive interactive experience and sense of participation.

Description

Technical Field

[0001] The present invention relates to the field of accident reconstruction technology, and in particular to an immersive safety accident reconstruction system and method based on digital twins. Background Art

[0002] In today's highly complex and diverse social and industrial environments, the frequency and severity of safety accidents are increasing, posing a significant threat to human life, property, and social stability. To effectively prevent and respond to safety accidents, in-depth research into their mechanisms, processes, and consequences is crucial. Digital twin technology, as an emerging digital tool, offers new insights and methods for reconstructing and analyzing safety accidents. Currently, existing technologies for reconstructing safety accidents have numerous limitations. Common methods rely on traditional two-dimensional images, text descriptions, or relatively crude three-dimensional models to represent the accident scene. These approaches clearly lack the realism and detail required to depict the accident scene.

[0003] However, existing systems mostly use relatively simple and general models for modeling scene facilities, and are unable to carry out refined and personalized design and construction for different types of scenes and specific facilities. Relying solely on a single modeling data source and lacking the multimodal data fusion of accident scene video texture features and three-dimensional remote sensing terrain, leads to distortion of the spatial topological relationship of scene facilities. This results in the inability to accurately display the unique structure, function, and operating status of facilities in different scenarios when recreating accidents, making it difficult to meet the needs of in-depth analysis and research on accidents. In addition, traditional methods use a splicing mode of static scenes + preset animations, which cannot achieve real-time dynamic coupling of personnel motion tracking data, accident development parameters, and scene models, and data synchronization errors are generally high. Existing VR systems only support fixed timeline playback and have not established a spatiotemporal slicing mechanism based on key moments in the accident process (such as the critical collision point and the peak energy release point), resulting in a high rate of missing immersive analysis of key causal factors.

[0004] Therefore, the present invention proposes an immersive safety accident reproduction system and method based on digital twins. Summary of the Invention

[0005] The present invention provides an immersive safety accident reconstruction system and method based on digital twins. The system comprises a scene rendering end that combines texture features of original accident site facilities, three-dimensional remote sensing terrain data, and original models of scene facilities to construct a refined original scene rendering model, providing a high-quality foundation for subsequent dynamic rendering. A dynamic fusion rendering end fuses multi-attribute characterization features, motion tracking data, and multi-dimensional accident development data with the original scene model for dynamic rendering, enabling the reconstruction model to more realistically and comprehensively reflect the accident situation. An immersive display end divides the dynamic model according to key moments in the accident process and presents it to users through virtual reality or augmented reality devices, allowing users to experience the accident reconstruction process in a more targeted and immersive way. This system can more accurately recreate the occurrence of safety accidents, providing an intuitive and effective means for accident analysis, prevention, and training. This system improves the accuracy, authenticity, and user experience of safety accident reconstruction, contributing to improved safety management and enhancing the user's immersive interactive experience and sense of participation. The present invention's multi-source data fusion architecture, dynamic rendering engine, and spatiotemporal slicing display mechanism overcome the shortcomings described in the background art and improve the accuracy of accident detail reconstruction.

[0006] The present invention provides an immersive safety accident reconstruction system based on digital twins, comprising:

[0007] The scene rendering end is used to build a refined original scene rendering model based on the texture features of the original accident scene facilities in the accident scene video, the 3D remote sensing terrain data of the target accident location, and the original scene facility models under each preset scene type;

[0008] The dynamic fusion rendering end is used to fuse the multi-attribute characterization features and motion tracking data of all people at the accident scene, as well as the multi-dimensional accident development representation data, with the refined original scene rendering model for dynamic rendering to obtain a dynamic model for reproducing safety accidents;

[0009] The immersive display terminal is used to divide the dynamic model of safety accident reconstruction based on all key moments of the accident process at the accident scene, obtain the local dynamic model of safety accident reconstruction corresponding to each key moment of the accident process, and display the local dynamic model of safety accident reconstruction corresponding to a single or multiple key moments of the accident process to the user through a virtual reality device or an augmented reality device.

[0010] Preferably, the scene rendering end includes:

[0011] A scene rough construction module is used to construct a rough original scene model based on the selection instructions input by the user and the original model of the scene facilities under each preset scene type;

[0012] The scene refinement rendering module is used to perform refinement on the rough original scene model based on the texture features of the original accident scene facilities in the accident scene video and the three-dimensional remote sensing terrain data of the target accident location to obtain a refined original scene rendering model.

[0013] Preferably, the dynamic fusion rendering end includes:

[0014] The human interaction analysis module is used to analyze the human interaction response data of all human scene interaction spatiotemporal nodes based on the multi-attribute characterization features and motion tracking data of all people at the accident scene, and generate the human interaction spatiotemporal recording thread of the accident scene;

[0015] The accident development analysis module is used to analyze the multi-dimensional accident development representation data based on the accident scene video, and generate the accident development spatiotemporal record thread of the accident scene based on the multi-dimensional accident development representation data;

[0016] The dynamic interaction analysis module is used to dynamically render the motion tracking data of all characters, the human interaction response data of all human-scene interaction spatiotemporal nodes, and the multi-dimensional accident development representation data based on the human interaction spatiotemporal recording thread at the accident scene and the accident development spatiotemporal recording thread, and the refined original scene rendering model to obtain a dynamic model for safety accident reproduction.

[0017] Preferably, the human interaction analysis module includes:

[0018] The character portrayal and motion tracking submodule is used to perform multi-attribute feature portrayal and motion tracking on all characters in the accident scene video, and obtain multi-attribute characterization features and motion tracking data of all characters in the accident scene;

[0019] A character dynamic model generation submodule is used to generate a three-dimensional motion tracking model of all characters at the accident scene based on the multi-attribute characterization features and motion tracking data of all characters at the accident scene;

[0020] The interaction node identification submodule is used to identify all human-scene interaction spatiotemporal nodes based on the motion tracking data of all people at the accident scene;

[0021] The interaction response analysis submodule is used to analyze the human interaction response data of all human scene interaction time and space nodes based on the accident scene video;

[0022] The interaction spatiotemporal record generation submodule is used to generate the human interaction spatiotemporal record thread of the accident scene based on the human interaction response data of all human scene interaction spatiotemporal nodes.

[0023] Preferably, the interactive node identification submodule includes:

[0024] a velocity and acceleration derivation unit, configured to determine the three-dimensional coordinates of each person at each moment based on the motion tracking data of all persons at the accident scene, and to determine the velocity and acceleration of each person in the three coordinate dimensions at each moment based on the three-dimensional coordinates of each person at each moment;

[0025] a rate of change vector construction unit, for calculating a velocity change rate vector and an acceleration change rate vector of each character at each moment based on the velocity and acceleration of each character in three coordinate dimensions at each moment;

[0026] an action change rate calculation unit, configured to calculate the comprehensive action change rate of each character at each moment based on the modulus of the velocity change rate vector and the modulus of the acceleration change rate vector of each character at each moment;

[0027] The interactive spatiotemporal node analysis unit is used to determine all artificial scene interactive spatiotemporal nodes of all characters based on the central coordinates of each facility in the refined original scene rendering model, the three-dimensional coordinates of each character at each moment, and the comprehensive change rate of each character's action at each moment.

[0028] Preferably, the interactive spatiotemporal node analysis unit includes:

[0029] a spatial correlation calculation subunit, configured to calculate the spatial correlation between each character and all facilities in the refined original scene rendering model at each moment based on the distance between the center coordinates of each facility in the refined original scene rendering model and the three-dimensional coordinates of each character at each moment;

[0030] a spatiotemporal correlation factor calculation subunit, configured to calculate the spatiotemporal correlation factor between each character and all facilities in the refined original scene rendering model at each moment based on the comprehensive change rate of each character's action at each moment and the spatial correlation between each character and all facilities in the refined original scene rendering model at each moment;

[0031] The interactive spatiotemporal node analysis subunit is used to determine all artificial scene interactive spatiotemporal nodes of all characters based on the spatiotemporal correlation factors between each character and all facilities in the refined original scene rendering model at each moment.

[0032] Preferably, the method of the interaction spatiotemporal node analysis subunit determining all artificial scene interaction spatiotemporal nodes of all characters based on the spatiotemporal correlation factor between each character and all facilities in the refined original scene rendering model at each moment includes:

[0033] All moments in which the spatiotemporal correlation factor between a single character and all facilities in the refined original scene rendering model at the corresponding moment is not less than the spatiotemporal correlation threshold are screened out, and regarded as all human-scene interaction time nodes of the corresponding character, and all human-scene interaction time nodes of the corresponding character and the three-dimensional coordinates of the corresponding character at all human-scene interaction time nodes are summarized, and regarded as all human-scene interaction spatiotemporal nodes of the corresponding character, until all human-scene interaction spatiotemporal nodes of all characters are determined.

[0034] Preferably, the immersive display terminal includes:

[0035] The accident process key point identification module is used to identify all the key moments of the accident process in the safety accident reconstruction dynamic model;

[0036] The accident process division module is used to divide the safety accident reconstruction dynamic model based on all the key moments of the accident process, and obtain the safety accident local reconstruction dynamic model corresponding to each key moment of the accident process;

[0037] The immersive display module is used to display the dynamic model of the local reproduction of the safety accident corresponding to the key moments of the accident process to the user through a virtual reality device or an augmented reality device based on the immersive interactive instructions input by the user.

[0038] Preferably, the accident process key point identification module includes:

[0039] A dynamic feature change rate analysis submodule is used to generate a multidimensional state vector at each moment of the accident scene based on the safety accident reconstruction dynamic model, and to generate a dynamic feature change rate vector at each moment of the accident scene based on the multidimensional state vector at each moment of the accident scene;

[0040] The accident key comprehensive index analysis submodule is used to analyze the correlation between different features at each moment of the accident scene based on the dynamic feature change rate vector at each moment of the accident scene, and determine the accident key comprehensive index at each moment of the accident scene based on the dynamic feature change rate vector at each moment of the accident scene and the correlation between different features;

[0041] The key moment screening submodule is used to screen out all moments in which the accident key point comprehensive index is not less than the accident key point comprehensive index threshold as all the key moments of the accident process in the safety accident reproduction dynamic model.

[0042] The present invention provides an immersive safety accident reconstruction method based on digital twins, which is applied to any of the above immersive safety accident reconstruction systems based on digital twins, comprising:

[0043] S1: Based on the texture features of the original accident scene facilities in the accident scene video, the 3D remote sensing terrain data of the target accident location, and the original model of the scene facilities under each preset scene type, a refined original scene rendering model is built;

[0044] S2: Dynamically render the scene by fusing the multi-attribute characterization features and motion tracking data of all characters at the accident scene, as well as the multi-dimensional accident development representation data, with the refined original scene rendering model to obtain a dynamic model for recreating the safety accident.

[0045] S3: Divide the dynamic model of safety accident reconstruction based on all accident process key points at the accident scene, obtain the local dynamic model of safety accident reconstruction corresponding to each accident process key point, and display the local dynamic model of safety accident reconstruction corresponding to a single or multiple accident process key points to the user through virtual reality equipment or augmented reality equipment.

[0046] The present invention offers the following advantages over existing technologies: the scene rendering end can combine the texture features of the original accident scene facilities, three-dimensional remote sensing terrain data, and the original model of the scene facilities to construct a refined original scene rendering model, providing a high-quality foundation for subsequent dynamic rendering. The dynamic fusion rendering end integrates the multi-attribute characterization features, motion tracking data, and multi-dimensional accident development data with the original scene model for dynamic rendering, making the reconstructed model more realistic and comprehensive in reflecting the accident situation. The immersive display end divides the dynamic model according to key moments in the accident process and presents it to the user through virtual reality or augmented reality devices, allowing the user to experience the accident reconstruction process in a more targeted and immersive way. It can more accurately reconstruct the occurrence of safety accidents, providing an intuitive and effective means for accident analysis, prevention, and training. It improves the accuracy, authenticity, and user experience of safety accident reconstruction, helps enhance safety management, and enhances the user's immersive interactive experience and sense of participation. The present invention's multi-source data fusion architecture, dynamic rendering engine, and spatiotemporal slicing display mechanism overcome the shortcomings described in the background art and improves the accuracy of accident detail reconstruction.

[0047] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in this application document.

[0048] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0050] Figure 1 Schematic diagram of an immersive safety accident reconstruction system based on digital twins in an embodiment of the present invention;

[0051] Figure 2 This is a schematic diagram of the execution logic of the immersive safety accident reconstruction system based on digital twins in an embodiment of the present invention;

[0052] Figure 3 This is a flow chart of the immersive safety accident reproduction method based on digital twins in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0054] Example 1:

[0055] The present invention provides an immersive safety accident reconstruction system based on digital twins. Figure 1 and Figure 2 ,include:

[0056] The scene rendering end is used to build a refined original scene rendering model based on the texture features of the original accident scene facilities in the accident scene video, the 3D remote sensing terrain data of the target accident location, and the original scene facility models under each preset scene type;

[0057] The dynamic fusion rendering end is used to fuse the multi-attribute characterization features and motion tracking data of all people at the accident scene, as well as the multi-dimensional accident development representation data, with the refined original scene rendering model for dynamic rendering to obtain a dynamic model for reproducing safety accidents;

[0058] The immersive display terminal is used to divide the dynamic model of safety accident reconstruction based on all key moments of the accident process at the accident scene, obtain the local dynamic model of safety accident reconstruction corresponding to each key moment of the accident process, and display the local dynamic model of safety accident reconstruction corresponding to a single or multiple key moments of the accident process to the user through a virtual reality device or an augmented reality device.

[0059] In this embodiment, the accident scene video refers to video data that records the actual situation at the scene of the safety accident, which includes various information such as the environmental conditions of the accident scene, the status of facilities, and the activities of people.

[0060] In this embodiment, the original accident scene facility texture features are the significant features such as surface texture, pattern, color, etc. presented by various facilities at the accident scene.

[0061] In this embodiment, the target accident location is the specific geographical location where the safety accident under study actually occurred.

[0062] In this embodiment, the three-dimensional remote sensing terrain data is three-dimensional data related to the terrain of the target accident location obtained through remote sensing technology, which can clearly reflect the actual conditions such as the ups and downs of the terrain.

[0063] In this embodiment, each preset scene type is a different type of accident scene that is preset in advance, such as a factory workshop scene, a road traffic accident scene, etc.

[0064] In this embodiment, the original model of the scene facilities under each preset scene type is the initial, unprocessed model of the scene facilities (such as buildings, structures, equipment and facilities inside and outside the factory, road greening, vehicles, tools, etc.) constructed for each pre-set scene type.

[0065] In this embodiment, the refined original scene rendering model is a scene model obtained by finely processing and rendering the texture features of the on-site facilities, terrain data and other elements based on the initially established model.

[0066] In this embodiment, the multi-attribute characterization feature of a person is a feature that describes in detail the attributes of multiple aspects of the person at the accident scene, such as age characteristics, gender characteristics, clothing characteristics, etc.

[0067] In this embodiment, the motion tracking data is data related to recording the motion of people at the accident scene.

[0068] In this embodiment, the multi-dimensional accident development characterization data is relevant data that describes the accident development process and status from multiple different aspects, such as the scale of the accident, the scope of impact, the severity, etc.

[0069] In this embodiment, the dynamic model for reproducing a safety accident is a model that is generated by fusing various types of relevant data at the accident scene and is capable of dynamically displaying the entire process of the accident.

[0070] In this embodiment, all key moments of the accident process at the accident scene are specific time points that have critical significance and important value in the entire process of the accident, such as the moment when the accident begins, the moment when a key event occurs, etc.

[0071] In this embodiment, the local dynamic model for reproducing the safety accident corresponding to each key moment of the accident process is a dynamic model generated for each key moment in the accident process and capable of independently displaying the local details of the accident at that specific moment.

[0072] The beneficial effects of the above technology are as follows: the scene rendering end can combine the texture features of the original accident scene facilities, three-dimensional remote sensing terrain data, and the original model of the scene facilities to build a refined original scene rendering model, providing a high-quality basic scene for subsequent dynamic rendering. The dynamic fusion rendering end integrates the multi-attribute character portrayal features, motion tracking data, and multi-dimensional accident development representation data with the original scene model for dynamic rendering, making the reconstruction model more realistic and comprehensive in reflecting the accident situation. The immersive display end divides the dynamic model according to the key moments of the accident process and displays it to the user through virtual reality or augmented reality devices, allowing the user to experience the accident reconstruction process in a more targeted and immersive way. It can more accurately reproduce the occurrence process of safety accidents, providing an intuitive and effective means for accident analysis, prevention, and training. It improves the accuracy, authenticity, and user experience of safety accident reconstruction, helps to improve the level of safety management, and enhances the user's immersive interactive experience and sense of participation. The multi-source data fusion architecture, dynamic rendering engine, and spatiotemporal slice display mechanism of the present invention solves the shortcomings described in the background technology and improves the accuracy of the restoration of accident details.

[0073] Example 2:

[0074] Based on Example 1, the scene rendering end includes:

[0075] A scene rough construction module is used to construct a rough original scene model based on the selection instructions input by the user and the original model of the scene facilities under each preset scene type;

[0076] The scene refinement rendering module is used to perform refinement on the rough original scene model based on the texture features of the original accident scene facilities in the accident scene video and the three-dimensional remote sensing terrain data of the target accident location to obtain a refined original scene rendering model.

[0077] In this embodiment, the selection instruction input by the user is an instruction given by the user according to his / her own needs to guide the operation of the system.

[0078] In this embodiment, based on the selection instructions input by the user and the original model of the scene facilities under each preset scene type, a rough original scene model is built. This is done in accordance with the instructions given by the user and combined with the initial facility models under various scene types set in advance to preliminarily construct a rough original scene model.

[0079] In this embodiment, the rough original scene model is finely rendered based on the texture features of the original accident scene facilities in the accident scene video and the three-dimensional remote sensing terrain data of the target accident location. The refined original scene rendering model is obtained by using the texture characteristics of the facilities in the accident scene video and the terrain data of the accident location to further carefully process and render the initially constructed scene model, thereby obtaining a more accurate and realistic scene rendering model.

[0080] The above technical solution has the following beneficial effects: The rough scene construction module quickly constructs a rough original scene model based on the user's input selection instructions and the original model, improving scene construction efficiency. The refined scene rendering module uses the texture characteristics of the accident scene facilities and 3D remote sensing terrain data to fine-tune the rough model, making the scene more realistic and accurate. This step-by-step construction approach improves the quality and detail of the scene while ensuring efficiency. It can provide a more realistic base model for subsequent dynamic fusion rendering. This improves the efficiency and effectiveness of scene rendering, helping to more realistically recreate the accident scene.

[0081] Example 3:

[0082] Based on Example 1, the dynamic fusion rendering end includes:

[0083] The human interaction analysis module is used to analyze the human interaction response data of all human scene interaction spatiotemporal nodes based on the multi-attribute characterization features and motion tracking data of all people at the accident scene, and generate the human interaction spatiotemporal recording thread of the accident scene;

[0084] The accident development analysis module is used to analyze the multi-dimensional accident development representation data based on the accident scene video, and generate the accident development spatiotemporal record thread of the accident scene based on the multi-dimensional accident development representation data;

[0085] The dynamic interaction analysis module is used to dynamically render the motion tracking data of all characters, the human interaction response data of all human-scene interaction spatiotemporal nodes, and the multi-dimensional accident development representation data based on the human interaction spatiotemporal recording thread at the accident scene and the accident development spatiotemporal recording thread, and the refined original scene rendering model to obtain a dynamic model for safety accident reproduction.

[0086] In this embodiment, the human-scene interaction spatiotemporal node refers to the specific time and space location where the human and scene facilities interact at the accident scene.

[0087] In this embodiment, the human interaction response data of the human scene interaction space-time node is the relevant data generated by the character's interactive behavior on the scene at the human scene interaction space-time node (for example, data indicating the change in the facility form after the task interacts with the scene).

[0088] In this embodiment, the spatiotemporal recording thread of human interaction at the accident scene is sequence data formed by continuously recording the temporal and spatial information of the interaction between people and scenes at the accident scene.

[0089] In this embodiment, the multi-dimensional accident development characterization data is obtained based on the accident scene video analysis, which is achieved by conducting an in-depth study of the accident scene video to obtain relevant data that can describe the accident development from multiple aspects.

[0090] In this embodiment, the spatiotemporal recording thread of the accident development of the accident scene is generated based on the multi-dimensional accident development representation data, which is to construct a record sequence reflecting the changes in the time and space of the accident development based on the acquired multi-dimensional accident development representation data.

[0091] In this embodiment, based on the human interaction spatiotemporal recording thread of the accident scene and the accident development spatiotemporal recording thread, the motion tracking data of all characters, the human interaction response data of all human-scene interaction spatiotemporal nodes, and the multi-dimensional accident development representation data are integrated with the refined original scene rendering model for dynamic rendering. The dynamic model for reproducing a safety accident is obtained by combining the spatiotemporal recording information of human interaction and accident development at the accident scene, together with the character motion tracking, human interaction response and multi-dimensional accident development data, with the refined original scene rendering model, and generating a dynamic model that can show the full picture of the accident through dynamic rendering.

[0092] The beneficial effects of the above technical solutions are as follows: the human interaction analysis module obtains human interaction response data and generates a human interaction spatiotemporal recording thread by analyzing the multi-attribute characterization features and motion tracking data of the characters, providing detailed information for a comprehensive understanding of the human factors in the accident. The accident development analysis module analyzes multi-dimensional accident development characterization data based on the accident scene video and generates an accident development spatiotemporal recording thread, which helps to accurately grasp the development process of the accident. The dynamic interaction analysis module combines the human interaction spatiotemporal recording thread and the accident development spatiotemporal recording thread for fusion dynamic rendering, making the dynamic model of safety accident reconstruction more in line with the actual situation. It can more comprehensively and realistically reflect the dynamic relationship between the characters and the environment in the accident, as well as the development of the accident. It improves the accuracy and authenticity of the dynamic model of safety accident reconstruction, providing a more valuable reference for accident analysis and prevention.

[0093] Example 4:

[0094] Based on Example 3, the human interaction analysis module includes:

[0095] The character portrayal and motion tracking submodule is used to perform multi-attribute feature portrayal and motion tracking on all characters in the accident scene video, and obtain multi-attribute characterization features and motion tracking data of all characters in the accident scene;

[0096] A character dynamic model generation submodule is used to generate a three-dimensional motion tracking model of all characters at the accident scene based on the multi-attribute characterization features and motion tracking data of all characters at the accident scene;

[0097] The interaction node identification submodule is used to identify all human-scene interaction spatiotemporal nodes based on the motion tracking data of all people at the accident scene;

[0098] The interaction response analysis submodule is used to analyze the human interaction response data of all human scene interaction time and space nodes based on the accident scene video;

[0099] The interaction spatiotemporal record generation submodule is used to generate the human interaction spatiotemporal record thread of the accident scene based on the human interaction response data of all human scene interaction spatiotemporal nodes.

[0100] In this embodiment, multi-attribute feature characterization and motion tracking are performed on all characters in the accident scene video. Obtaining multi-attribute characterization features and motion tracking data for all characters in the accident scene means carefully describing each character in the accident scene video in terms of multiple attributes, and tracking their motions at the same time, thereby obtaining multi-faceted feature descriptions of these characters and relevant data on their motions.

[0101] In this embodiment, generating a three-dimensional motion tracking model for all characters at the accident scene based on the multi-attribute characterization features and motion tracking data of all characters at the accident scene is to use the multi-attribute features and motion tracking data of the characters obtained at the accident scene to construct a model that can display the three-dimensional motion of these characters.

[0102] In this embodiment, analyzing the human interaction response data for all spatiotemporal nodes of human-scene interaction based on the accident scene video involves studying the accident scene video to obtain relevant data related to the reactions of scene facilities caused by the interactive actions performed by the characters at specific spatiotemporal nodes of the characters-scene interaction. To achieve this, the video must first be preprocessed, its resolution adjusted to an appropriate state, and its frame rate unified for subsequent processing. Next, a deep learning target detection algorithm such as YOLO and FasterR-CNN is used to detect the characters and scene facilities, and their motion trajectories are then obtained using a multi-target tracking algorithm such as DeepSORT. Keyframes are then extracted, and the interaction spatiotemporal nodes are determined based on the position and posture changes of the characters and scene facilities in the keyframes and adjacent frames. Finally, at these nodes, the character interaction actions are analyzed using a posture estimation algorithm, while the displacement, deformation, and other reaction data of the scene facilities under the interaction actions are observed and recorded.

[0103] In this embodiment, the human interaction spatiotemporal recording thread of the accident scene is generated based on the human interaction response data of all human scene interaction spatiotemporal nodes, which is to compile a record sequence that can reflect the changes in time and space of human interaction at the accident scene based on the human interaction response data obtained on the human scene interaction spatiotemporal nodes.

[0104] The beneficial effects of the above technical solutions are as follows: the character portrayal and motion tracking submodule performs multi-attribute feature portrayal and motion tracking on characters, providing a rich data foundation for subsequent analysis. The character dynamic model generation submodule generates a three-dimensional motion tracking model, making the character's motion performance more intuitive and clear. The interaction node identification submodule can accurately identify the spatiotemporal nodes of human-scene interactions and capture key interaction moments. The interaction response analysis submodule analyzes the interaction response data to gain an in-depth understanding of the character's behavior at the interaction nodes. The interaction spatiotemporal record generation submodule generates a human interaction spatiotemporal record thread, providing important clues and basis for the overall dynamic fusion rendering. This improves the accuracy and comprehensiveness of human interaction analysis and helps to more realistically reproduce the human factors in accidents.

[0105] Example 5:

[0106] Based on Example 4, the interactive node identification submodule includes:

[0107] a velocity and acceleration derivation unit, configured to determine the three-dimensional coordinates of each person at each moment based on the motion tracking data of all persons at the accident scene, and to determine the velocity and acceleration of each person in the three coordinate dimensions at each moment based on the three-dimensional coordinates of each person at each moment;

[0108] a rate of change vector construction unit, for calculating a velocity change rate vector and an acceleration change rate vector of each character at each moment based on the velocity and acceleration of each character in three coordinate dimensions at each moment;

[0109] an action change rate calculation unit, configured to calculate the comprehensive action change rate of each character at each moment based on the modulus of the velocity change rate vector and the modulus of the acceleration change rate vector of each character at each moment;

[0110] The interactive spatiotemporal node analysis unit is used to determine all artificial scene interactive spatiotemporal nodes of all characters based on the central coordinates of each facility in the refined original scene rendering model, the three-dimensional coordinates of each character at each moment, and the comprehensive change rate of each character's action at each moment.

[0111] In this embodiment, determining the speed and acceleration of each character in the three coordinate dimensions at each moment based on the three-dimensional coordinates of each character at each moment is to calculate the character's movement speed and the speed change rate (acceleration) in the three coordinate directions of X, Y, and Z at this moment based on the spatial position coordinates of the character at each specific time point using relevant mathematical methods.

[0112] In this embodiment, the velocity change rate vector and acceleration change rate vector of each character at each moment are calculated based on the velocity and acceleration of each character in the three coordinate dimensions at each moment. This is to use the velocity and acceleration data of the character in the three directions at each moment, and further obtain a vector that can describe the speed and acceleration change speed through mathematical operations. Among them, the velocity change rate vector is composed of the velocity change rate in the X, Y, and Z axis directions at a single moment, and the acceleration change rate vector is composed of the acceleration change rate in the X, Y, and Z axis directions at a single moment.

[0113] In this embodiment, the comprehensive change rate of each character's movement at each moment is calculated based on the modulus of the velocity change rate vector and the modulus of the acceleration change rate vector of each character at each moment. This is done by combining the magnitude of the velocity change rate vector and the magnitude of the acceleration change rate vector at each moment, thereby obtaining a numerical value that reflects the overall intensity of the change in the character's movement at that moment. For example, a weighted summation method is used to calculate the product of the modulus of each character's velocity change rate vector and the corresponding weight coefficient, and the product of the modulus of the acceleration change rate vector and the corresponding weight coefficient at each moment, and the sum of these two products is used as the comprehensive change rate of each character's movement at the corresponding moment, where the sum of the two weight coefficients is 1.

[0114] The beneficial effects of the above technical solution are as follows: the speed and acceleration derivation unit calculates the speed and acceleration by determining the three-dimensional coordinates of the character, providing basic motion parameters for subsequent analysis. The change rate vector construction unit calculates the speed change rate vector and the acceleration change rate vector to more comprehensively describe the changes in the character's motion. The action change rate calculation unit derives the comprehensive action change rate, which can quantitatively measure the intensity and degree of change of the character's action. The interactive spatiotemporal node analysis unit comprehensively considers the scene facility coordinates, the character's three-dimensional coordinates and the comprehensive action change rate to determine the interactive spatiotemporal nodes, thereby improving the accuracy and reliability of recognition. It can more accurately identify the artificial scene interactive spatiotemporal nodes, providing key time points and position information for subsequent interactive response analysis and dynamic fusion rendering.

[0115] Example 6:

[0116] Based on Example 5, the interactive spatiotemporal node analysis unit includes:

[0117] The spatial correlation calculation subunit is used to calculate the spatial correlation between each character and all facilities in the refined original scene rendering model at each moment based on the distance between the center coordinates of each facility in the refined original scene rendering model and the three-dimensional coordinates of each character at each moment (which is the sum of the weighted inverses of the distances between the character and all facilities):

[0118]

[0119] Where S(t) is the spatial correlation between the currently calculated single person at time t and all facilities in the refined original scene rendering model, m is the total number of facilities in the refined original scene rendering model, and γ j is the weight coefficient of the jth facility in the refined original scene rendering model, d j (t) is the distance between the center coordinates of the jth facility in the refined original scene rendering model and the currently calculated three-dimensional coordinates of the single character at time t,

[0120] The spatiotemporal correlation factor calculation subunit is used to calculate the spatiotemporal correlation factor (combining the action change in the time dimension and the correlation in the spatial dimension) between each character and all facilities in the refined original scene rendering model at each moment based on the comprehensive change rate of each character's action at each moment and the spatial correlation between each character and all facilities in the refined original scene rendering model at each moment:

[0121] C(t)=R(t)*S(t)

[0122] Where C(t) is the spatiotemporal correlation factor between each character at time t and all facilities in the refined original scene rendering model, and R(t) is the comprehensive change rate of each character's action at time t.

[0123] The interactive spatiotemporal node analysis subunit is used to determine all artificial scene interactive spatiotemporal nodes of all characters based on the spatiotemporal correlation factors between each character and all facilities in the refined original scene rendering model at each moment.

[0124] In this embodiment, the spatial correlation between each character and all facilities in the refined original scene rendering model at each moment is a metric used to measure the degree of closeness of the spatial relationship between a single character and all facilities in the refined original scene rendering model at a specific moment.

[0125] In this embodiment, the spatiotemporal correlation factor between each character and all facilities in the refined original scene rendering model at each moment is an indicator that comprehensively considers the character's actions at a specific moment and its spatial relationship with all facilities, and is used to more comprehensively describe the degree of comprehensive temporal and spatial correlation between the character and the scene facilities at this moment.

[0126] The beneficial effects of the above technical solution are as follows: the spatial correlation calculation subunit calculates the distance between people and facilities to obtain the spatial correlation, quantifying the positional relationship between people and environmental facilities. The spatiotemporal correlation factor calculation subunit comprehensively considers the comprehensive change rate of actions and spatial correlation to more comprehensively evaluate the degree of temporal and spatial correlation between people and facilities. The interactive spatiotemporal node analysis subunit determines the interactive spatiotemporal nodes based on the spatiotemporal correlation factors, making the determination of nodes more scientific and accurate. It can more accurately capture the moments and locations of important temporal and spatial interactions between people and scene facilities. By extracting multi-dimensional features from motion tracking data, quantifying the spatial correlation between people and scene facilities, calculating the spatiotemporal correlation factors, and making judgments based on set thresholds, the function of accurately identifying all human-scene interactive spatiotemporal nodes from motion tracking data is achieved. This improves the accuracy and reliability of the interactive spatiotemporal node analysis, providing strong support for more realistic reproduction of human interactions in accidents.

[0127] Example 7:

[0128] On the basis of Example 6, the method of the interactive spatiotemporal node analysis subunit determining all artificial scene interactive spatiotemporal nodes of all characters based on the spatiotemporal correlation factor between each character and all facilities in the refined original scene rendering model at each moment includes:

[0129] All moments in which the spatiotemporal correlation factor between a single character and all facilities in the refined original scene rendering model at the corresponding moment is not less than the spatiotemporal correlation threshold are screened out, and regarded as all human-scene interaction time nodes of the corresponding character, and all human-scene interaction time nodes of the corresponding character and the three-dimensional coordinates of the corresponding character at all human-scene interaction time nodes are summarized, and regarded as all human-scene interaction spatiotemporal nodes of the corresponding character, until all human-scene interaction spatiotemporal nodes of all characters are determined.

[0130] In this embodiment, the spatiotemporal correlation threshold is a pre-set standard value for determining whether the spatiotemporal correlation between a person and a scene facility reaches a significant level. If the spatiotemporal correlation factor between a person and a facility exceeds this threshold, it is considered to have significant interaction significance.

[0131] In this embodiment, the human-scene interaction time node refers to a specific time point at the accident scene when a person and the scene have a significant interaction.

[0132] The beneficial effects of the above technical solution are as follows: by setting a spatiotemporal correlation threshold, moments with significant interactive significance can be accurately screened out, avoiding interference from irrelevant or minor interactive moments. The three-dimensional coordinates of the characters at the interaction time nodes are aggregated together to fully determine the spatiotemporal nodes of human-induced scene interactions, providing precise location information for subsequent analysis and rendering. This determination method has clear standards and operability, and improves the accuracy and consistency of the determination of interactive spatiotemporal nodes. It can effectively highlight key human interaction moments and locations, and help to more accurately reproduce important interactive scenes in accidents. It improves the scientific nature and effectiveness of the determination of spatiotemporal nodes of human-induced scene interactions, and provides more valuable interaction data for the safety accident reconstruction system.

[0133] Example 8:

[0134] Based on Example 1, the immersive display terminal includes:

[0135] The accident process key point identification module is used to identify all the key moments of the accident process in the safety accident reconstruction dynamic model;

[0136] The accident process division module is used to divide the safety accident reconstruction dynamic model based on all the key moments of the accident process, and obtain the safety accident local reconstruction dynamic model corresponding to each key moment of the accident process;

[0137] The immersive display module is used to display the dynamic model of the local reproduction of the safety accident corresponding to the key moments of the accident process to the user through a virtual reality device or an augmented reality device based on the immersive interactive instructions input by the user.

[0138] In this embodiment, the dynamic model of safety accident reconstruction is divided based on all key moments of the accident process, and the local dynamic model of safety accident reconstruction corresponding to each key moment of the accident process is obtained. This is based on all key time points in the accident process, and the dynamic model of the entire safety accident reconstruction is decomposed into multiple parts. Each part is a local dynamic model corresponding to a key moment of the accident, and can separately display the specific circumstances of the accident at that moment.

[0139] In this embodiment, based on the immersive interactive instructions input by the user and through a virtual reality device or an augmented reality device, a dynamic model of the local reproduction of the safety accident corresponding to the key moment of the accident process is displayed to the user. This is in accordance with the instructions given by the user who wishes to have a deeper understanding and experience, and using virtual reality or augmented reality technology, the dynamic model of the local reproduction of the safety accident corresponding to the key moment of the accident process specified by the user is presented to the user, allowing the user to have an immersive experience.

[0140] The beneficial effects of the above technical solutions are as follows: the accident process key point identification module can accurately identify the key moments in the safety accident reconstruction dynamic model, providing a basis for subsequent model division. The accident process division module divides the model according to the key moments, making each local reconstruction dynamic model more targeted and independent. The immersive display module displays according to the user's immersive interactive instructions, enhancing the user's sense of participation and experience. It can meet the user's attention and in-depth understanding of the key points of a specific accident process, and improve the flexibility and pertinence of the display. It improves the accuracy and user-friendliness of the safety accident reconstruction display, which helps to better play the role of the system in accident analysis and training.

[0141] Example 9:

[0142] Based on Example 8, the accident process key point identification module includes:

[0143] A dynamic feature change rate analysis submodule is used to generate a multidimensional state vector at each moment of the accident scene based on the safety accident reconstruction dynamic model, and to generate a dynamic feature change rate vector at each moment of the accident scene based on the multidimensional state vector at each moment of the accident scene;

[0144] The accident key comprehensive index analysis submodule is used to analyze the correlation between different features at each moment of the accident scene based on the dynamic feature change rate vector at each moment of the accident scene, and determine the accident key comprehensive index at each moment of the accident scene based on the dynamic feature change rate vector at each moment of the accident scene and the correlation between different features;

[0145] The key moment screening submodule is used to screen out all moments in which the accident key point comprehensive index is not less than the accident key point comprehensive index threshold as all the key moments of the accident process in the safety accident reproduction dynamic model.

[0146] In this embodiment, the dynamic model for reconstructing safety accidents generates a multi-dimensional state vector for each moment of the accident scene. This is to create a vector that can describe its state from multiple aspects for each specific time point of the accident scene through the dynamic model for reconstructing safety accidents. For example, for time t, a multi-dimensional feature vector X(t) = [x1(t), x2(t), ..., x n (t)], where x i (t) represents the value of the i-th feature dimension at time t. For example, x1(t) can be the displacement of the key object in the accident, x2(t) is its velocity, x3(t) is the acceleration, etc.

[0147] In this embodiment, the dynamic characteristic change rate vector of each moment of the accident scene is generated based on the multi-dimensional state vector of each moment. The vector that can reflect the speed of change of these states is calculated based on the multi-dimensional state vector of each moment:

[0148]

[0149] Where Δt is a very small time interval used to calculate the rate of change of features in a short period of time. This vector reflects the speed of change of each feature dimension around time t.

[0150] In this embodiment, the correlation between different features of the accident scene at each moment is analyzed based on the dynamic feature change rate vector at each moment. This is to use the dynamic feature change rate vector at each moment to study the mutual correlation and influence relationship between various features of the accident scene at that moment. For example, the feature change correlation matrix C(t) is introduced, and its element C ij (t) represents the correlation between the i-th feature change rate and the j-th feature change rate at time t:

[0151]

[0152] where R i (t) and R j (t) are the i-th and j-th elements in R(t), respectively. i (t k ) and R j (t k ) are R(t k ), the i-th and j-th elements in and is their average value within a time window N, t kFor the kth moment within a time window N, the correlation matrix helps to discover the synergistic or abnormal relationship between feature changes.

[0153] In this embodiment, the comprehensive index of the key points of the accident at each moment is determined based on the dynamic feature change rate vector at each moment and the correlation between different features. This is to calculate a comprehensive index that can fully reflect the importance of the moment in the accident by combining the dynamic feature change rate vector at each moment and the correlation between different features. For example, the comprehensive index of the key points of the accident at time t is defined as:

[0154]

[0155] Where, α i is the weight of the characteristic change rate, reflecting its importance in judging the key points of the accident. β ij is the weight of the correlation matrix element, reflecting the contribution of the correlation of different feature changes to the judgment of the key points of the accident.

[0156] In this embodiment, the accident key point comprehensive index threshold is a pre-set standard value used to judge whether the importance of each moment at the accident scene reaches a critical level. If the accident key point comprehensive index at a certain moment exceeds this threshold, this moment is considered to be a critical point moment of the accident.

[0157] The beneficial effects of the above technical solution are as follows: the dynamic feature change rate analysis submodule generates a multidimensional state vector and a dynamic feature change rate vector, providing basic data for subsequent analysis. The accident key point comprehensive index analysis submodule determines the comprehensive index by analyzing the correlation, and can comprehensively evaluate the importance of each moment. The key moment screening submodule filters out the key moments by setting thresholds, ensuring the accuracy and effectiveness of the identification. It can accurately identify the key point moments in the dynamic model of safety accident reproduction, and provide key nodes for subsequent model division and display. It improves the accuracy and reliability of the identification of key points in the accident process, and helps to display important accident processes in a more targeted manner.

[0158] Example 10:

[0159] The present invention provides an immersive safety accident reproduction method based on digital twins, which is applied to any one of the immersive safety accident reproduction systems based on digital twins in Examples 1 to 9, with reference to Figure 3 ,include:

[0160] S1: Based on the texture features of the original accident scene facilities in the accident scene video, the 3D remote sensing terrain data of the target accident location, and the original model of the scene facilities under each preset scene type, a refined original scene rendering model is built;

[0161] S2: Dynamically render the scene by fusing the multi-attribute characterization features and motion tracking data of all characters at the accident scene, as well as the multi-dimensional accident development representation data, with the refined original scene rendering model to obtain a dynamic model for recreating the safety accident.

[0162] S3: Divide the dynamic model of safety accident reconstruction based on all accident process key points at the accident scene, obtain the local dynamic model of safety accident reconstruction corresponding to each accident process key point, and display the local dynamic model of safety accident reconstruction corresponding to a single or multiple accident process key points to the user through virtual reality equipment or augmented reality equipment.

[0163] The beneficial effects of the above technology are as follows: Step S1 can combine the texture features of the original accident scene facilities, three-dimensional remote sensing terrain data and the original model of the scene facilities to build a refined original scene rendering model, providing a high-quality basic scene for subsequent dynamic rendering. Step S2 integrates the multi-attribute characterization features, motion tracking data and multi-dimensional accident development characterization data with the original scene model for dynamic rendering, so that the reproduction model reflects the accident situation more realistically and comprehensively. Step S3 divides the dynamic model according to the key moments of the accident process and displays it to the user through virtual reality or augmented reality devices, allowing the user to experience the accident reproduction process more targeted and immersive. It can more accurately reproduce the occurrence process of safety accidents and provide an intuitive and effective means for accident analysis, prevention and training. It improves the accuracy, authenticity and user experience of safety accident reproduction, helps to improve the level of safety management, and enhances the user's immersive interactive experience and sense of participation. The multi-source data fusion architecture, dynamic rendering engine and spatiotemporal slice display mechanism of the present invention solve the defects described in the background technology and improve the accuracy of the restoration of accident details.

[0164] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An immersive safety accident reconstruction system based on digital twins, characterized by: include: The scene rendering end is used to build a refined original scene rendering model based on the texture features of the original accident scene facilities in the accident scene video, the 3D remote sensing terrain data of the target accident location, and the original scene facility models under each preset scene type; The dynamic fusion rendering end is used to fuse the multi-attribute characterization features and motion tracking data of all people at the accident scene, as well as the multi-dimensional accident development representation data, with the refined original scene rendering model for dynamic rendering to obtain a dynamic model for reproducing safety accidents; The immersive display terminal is used to divide the safety accident reconstruction dynamic model based on all key moments of the accident process at the accident scene, obtain the safety accident local reconstruction dynamic model corresponding to each key moment of the accident process, and display the safety accident local reconstruction dynamic model corresponding to a single or multiple key moments of the accident process to the user through a virtual reality device or an augmented reality device, including: The accident process key point identification module is used to identify all the key moments of the accident process in the safety accident reconstruction dynamic model, including: The dynamic characteristic change rate analysis submodule is used to generate a multi-dimensional state vector at each moment of the accident scene based on the safety accident reconstruction dynamic model, that is, for the moment , define the multidimensional state vector ,in, Indicates the feature dimension at time The value of , and based on the multi-dimensional state vector at each moment of the accident scene, the dynamic characteristic change rate vector at each moment of the accident scene is generated : in, It is a very small time interval used to calculate the rate of change of features in a short period of time. The dynamic feature change rate vector reflects the change of each feature dimension at the moment How fast the neighborhood changes; The accident key comprehensive index analysis submodule is used to analyze the correlation between different features at each moment of the accident scene based on the dynamic feature change rate vector at each moment of the accident scene: in, Indicates time No. The rate of change of the characteristic The correlation between the change rates of the features, and They are The Hedi elements, and They are The Hedi elements, and They are in a time window The average value within A time window The first a moment; Based on the dynamic feature change rate vector at each moment of the accident scene and the correlation between different features, the comprehensive index of the accident key points at each moment of the accident scene is determined: Where, For the moment Comprehensive indicators of accident key points, It is The weight of the characteristic change rate reflects its importance to the judgment of the key points of the accident. ; is the weight of the correlation matrix element, reflecting the contribution of the correlation of different feature changes to the judgment of the key points of the accident. ; The key moment screening submodule is used to screen out all moments in which the accident key point comprehensive index is not less than the accident key point comprehensive index threshold as all the key moments of the accident process in the safety accident reconstruction dynamic model; The accident process division module is used to divide the safety accident reconstruction dynamic model based on all the key moments of the accident process, and obtain the safety accident local reconstruction dynamic model corresponding to each key moment of the accident process; The immersive display module is used to display the dynamic model of the local reproduction of the safety accident corresponding to the key moments of the accident process to the user through a virtual reality device or an augmented reality device based on the immersive interactive instructions input by the user.

2. The immersive safety accident reconstruction system based on digital twin according to claim 1 is characterized in that: Scene rendering end, including: A scene rough construction module is used to construct a rough original scene model based on the selection instructions input by the user and the original model of the scene facilities under each preset scene type; The scene refinement rendering module is used to perform refinement on the rough original scene model based on the texture features of the original accident scene facilities in the accident scene video and the three-dimensional remote sensing terrain data of the target accident location to obtain a refined original scene rendering model.

3. The immersive safety accident reconstruction system based on digital twin according to claim 1 is characterized in that: Dynamic fusion rendering end, including: The human interaction analysis module is used to analyze the human interaction response data of all human scene interaction spatiotemporal nodes based on the multi-attribute characterization features and motion tracking data of all people at the accident scene, and generate the human interaction spatiotemporal recording thread of the accident scene; The accident development analysis module is used to analyze the multi-dimensional accident development representation data based on the accident scene video, and generate the accident development spatiotemporal record thread of the accident scene based on the multi-dimensional accident development representation data; The dynamic interaction analysis module is used to dynamically render the motion tracking data of all characters, the human interaction response data of all human-scene interaction spatiotemporal nodes, and the multi-dimensional accident development representation data based on the human interaction spatiotemporal recording thread at the accident scene and the accident development spatiotemporal recording thread, and the refined original scene rendering model to obtain a dynamic model for safety accident reproduction.

4. The immersive safety accident reconstruction system based on digital twin according to claim 3 is characterized in that: Human interaction analysis module, including: The character portrayal and motion tracking submodule is used to perform multi-attribute feature portrayal and motion tracking on all characters in the accident scene video, and obtain multi-attribute characterization features and motion tracking data of all characters in the accident scene; A character dynamic model generation submodule is used to generate a three-dimensional motion tracking model of all characters at the accident scene based on the multi-attribute characterization features and motion tracking data of all characters at the accident scene; The interaction node identification submodule is used to identify all human-scene interaction spatiotemporal nodes based on the motion tracking data of all people at the accident scene; The interaction response analysis submodule is used to analyze the human interaction response data of all human scene interaction time and space nodes based on the accident scene video; The interaction spatiotemporal record generation submodule is used to generate the human interaction spatiotemporal record thread of the accident scene based on the human interaction response data of all human scene interaction spatiotemporal nodes.

5. The immersive safety accident reconstruction system based on digital twin according to claim 4 is characterized in that: The interactive node identification submodule includes: a velocity and acceleration derivation unit, configured to determine the three-dimensional coordinates of each person at each moment based on the motion tracking data of all persons at the accident scene, and to determine the velocity and acceleration of each person in the three coordinate dimensions at each moment based on the three-dimensional coordinates of each person at each moment; a rate of change vector construction unit, for calculating a velocity change rate vector and an acceleration change rate vector of each character at each moment based on the velocity and acceleration of each character in three coordinate dimensions at each moment; an action change rate calculation unit, configured to calculate the comprehensive action change rate of each character at each moment based on the modulus of the velocity change rate vector and the modulus of the acceleration change rate vector of each character at each moment; The interactive spatiotemporal node analysis unit is used to determine all artificial scene interactive spatiotemporal nodes of all characters based on the center coordinates of each facility in the refined original scene rendering model, the three-dimensional coordinates of each character at each moment, and the comprehensive change rate of each character's action at each moment, including: a spatial correlation calculation subunit, configured to calculate the spatial correlation between each character and all facilities in the refined original scene rendering model at each moment based on the distance between the center coordinates of each facility in the refined original scene rendering model and the three-dimensional coordinates of each character at each moment, wherein the spatial correlation is the sum of the weighted inverses of the distances between the character and all facilities; The spatiotemporal correlation factor calculation subunit is used to calculate the spatiotemporal correlation factor between each character and all facilities in the refined original scene rendering model at each moment based on the comprehensive change rate of each character's actions at each moment and the spatial correlation between each character and all facilities in the refined original scene rendering model at each moment: Where, For each character at the moment The spatiotemporal correlation factors between all facilities in the refined original scene rendering model, For each character at the moment The comprehensive rate of change of action, is the spatial correlation, which is the sum of the weighted inverses of the distances between the person and all facilities; The interactive spatiotemporal node analysis subunit is used to determine all artificial scene interactive spatiotemporal nodes of all characters based on the spatiotemporal correlation factors between each character and all facilities in the refined original scene rendering model at each moment.

6. The immersive safety accident reconstruction system based on digital twin according to claim 5 is characterized in that: The interactive spatiotemporal node analysis subunit determines the method of all artificial scene interactive spatiotemporal nodes of all characters based on the spatiotemporal correlation factors between each character and all facilities in the refined original scene rendering model at each moment, including: All moments in which the spatiotemporal correlation factor between a single character and all facilities in the refined original scene rendering model at the corresponding moment is not less than the spatiotemporal correlation threshold are screened out, and regarded as all human-scene interaction time nodes of the corresponding character, and all human-scene interaction time nodes of the corresponding character and the three-dimensional coordinates of the corresponding character at all human-scene interaction time nodes are summarized, and regarded as all human-scene interaction spatiotemporal nodes of the corresponding character, until all human-scene interaction spatiotemporal nodes of all characters are determined.

7. An immersive safety accident reconstruction method based on digital twins, characterized in that: The immersive safety accident reconstruction system based on digital twins as described in any one of claims 1 to 6 comprises: S1: Based on the texture features of the original accident scene facilities in the accident scene video, the 3D remote sensing terrain data of the target accident location, and the original model of the scene facilities under each preset scene type, a refined original scene rendering model is built; S2: Dynamically render the scene by fusing the multi-attribute characterization features and motion tracking data of all characters at the accident scene, as well as the multi-dimensional accident development representation data, with the refined original scene rendering model to obtain a dynamic model for recreating the safety accident. S3: Divide the dynamic model of safety accident reconstruction based on all accident process key points at the accident scene, obtain the local dynamic model of safety accident reconstruction corresponding to each accident process key point, and display the local dynamic model of safety accident reconstruction corresponding to a single or multiple accident process key points to the user through virtual reality equipment or augmented reality equipment.

Images