Digital-twin-based two-and-three-dimensional integrated geographic scene intelligent display system

Abstract

The application discloses a two-three-dimensional integrated geographic scene intelligent display system based on digital twinning, and relates to the technical field of digital twinning.The system comprises a two-three-dimensional scene construction module, a knowledge graph semantic module, an edge scene analysis and cooperation module and an intelligent display interaction module.The application establishes an associated mapping mechanism of geographic data and knowledge graph entities by constructing a four-module cooperative architecture of two-three-dimensional scene construction, knowledge graph semantics, edge scene analysis and cooperation and intelligent display interaction, realizes two-three-dimensional scene cooperative modeling and seamless switching, dynamically adjusts scene rendering precision in combination with edge analysis resource state, realizes semantic integrated construction and adaptive rendering of two-three-dimensional geographic scenes, and therefore can solve the technical problems of low interaction precision, mismatching of rendering precision and resource occupation and influence on system operation efficiency caused by the fragmentation of two-three-dimensional geographic scenes and the lack of semantic identification of scene elements in the prior art.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to a two- or three-dimensional integrated intelligent display system for geographic scenes based on digital twins. Background Technology

[0002] The integrated 2D and 3D geographic scene intelligent display system restores the geographic environment and displays it intelligently and visually by merging 2D and 3D data from the same source and multiple sources. It supports functions such as spatial analysis, situation mapping, and simulation, and is compatible with multi-terminal interaction, thereby improving management efficiency, response speed, and the scientific nature of decision-making.

[0003] Currently, integrated 2D and 3D geographic scene intelligent display systems generally suffer from problems such as fragmented 2D and 3D geographic scenes, lack of semantic identification for scene elements, and geographic knowledge graphs that are mostly general-purpose and difficult to adapt to specific scene requirements and are updated late. At the same time, edge analysis deployment is cumbersome, analysis tasks are redundant, analysis results are disconnected from semantic information, and display interaction modes are monotonous. These issues affect the accuracy of system interaction and operational efficiency, resulting in insufficient semantic support and inaccurate user perception of scene changes and related information.

[0004] Therefore, a digital twin-based integrated 2D / 3D geographic scene intelligent display system is proposed to solve the above problems. Summary of the Invention

[0005] The main objective of this invention is to provide a digital twin-based integrated 2D / 3D geographic scene intelligent display system to solve the problems mentioned in the background above.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a digital twin-based integrated 2D / 3D geographic scene intelligent display system, the system comprising:

[0007] The knowledge graph semantic module is used to construct a geographic knowledge graph system, perform semantic queries and related displays, and dynamically and incrementally update the geographic knowledge graph system; the geographic knowledge graph system includes knowledge graph entities identified by semantic tags and semantic relationships between geographic entities;

[0008] The 2D / 3D scene construction module performs collaborative modeling of 2D / 3D scenes based on semantic relationships, and dynamically adjusts the scene rendering accuracy based on the resource usage status fed back by the edge scene analysis and collaboration module.

[0009] The edge scene analysis collaboration module is used to build a lightweight edge analysis deployment system, perform real-time analysis of 2D and 3D twin scene data, establish semantic association links between edge analysis results and knowledge graph, generate fusion analysis data and push it to the knowledge graph semantic module, and collect its own resource occupancy status and push it to the 2D and 3D scene construction module.

[0010] The intelligent display and interaction module is used to receive and integrate the fusion analysis data pushed by the edge scene analysis and collaboration module, respond to the user's geographic semantic query request, and display the related data in the two-dimensional and three-dimensional twin scene based on the query results returned by the knowledge graph semantic module.

[0011] Preferably, the semantic module of the knowledge graph is specifically used for: when constructing a geographic knowledge graph system.

[0012] Collect geographic data, including terrain data, image data, and vector data for constructing 2D and 3D twin scenes;

[0013] Geographic entities and their attribute information are extracted from geographic data. A geographic entity is a data unit that expresses a geographic phenomenon. Specifically, a geographic entity is a data unit that expresses a geographic phenomenon that is formed naturally, artificially constructed, or administratively managed.

[0014] The geographic entity attribute information is represented as a set of feature data of the geographic entity, specifically divided into general attribute information and classification attribute information;

[0015] The extracted geographic entities are classified, semantic relationships between different geographic entities are established, and a unique semantic identifier is assigned to each geographic entity.

[0016] Based on the classified geographic entities, geographic entity attribute information, and semantic relationships, a geographic knowledge graph system is constructed.

[0017] The knowledge graph entities correspond one-to-one with geographic entities and semantic identifiers.

[0018] Preferably, when performing semantic queries and related display, the knowledge graph semantic module is specifically used for:

[0019] Based on the semantic relationships in the geographic knowledge graph system, the geographic entities associated with the semantic attributes of scene elements are queried. The scene elements are two-dimensional vector primitives or three-dimensional models that visualize geographic entities in a two-dimensional or three-dimensional twin scene.

[0020] The location and attribute tagging information of the scene elements corresponding to the associated geographic entities in the associated query results will be pushed to the intelligent display and interaction module.

[0021] When the intelligent display interaction module triggers the selection operation of the scene element corresponding to the geographic entity, it extracts the related entity information and historical related data of the geographic entity.

[0022] The associated entity information is the attribute information of other geographic entities that have a semantic relationship with the selected geographic entity, and the historical association data is the record data of the historical association relationship between the selected geographic entity and the associated geographic entities.

[0023] The extracted related entity information and historical related data are pushed to the intelligent display and interaction module.

[0024] Preferably, when dynamically and incrementally updating the geographic knowledge graph system, the semantic module of the knowledge graph is specifically used for:

[0025] Real-time monitoring of updated data, including fusion analysis data pushed by the edge scene analysis collaboration module and updated geographic data;

[0026] The monitored updated data is used to identify relationships, determine newly added geographic entities, newly added geographic entity attribute information and newly added semantic relationships, or identify changes in existing geographic entity attribute information and semantic relationships.

[0027] The identified new or changed content is reviewed, and once approved, it is added to the geographic knowledge graph system to complete the dynamic incremental update of the graph.

[0028] Preferably, the 2D / 3D scene construction module, when performing collaborative 2D / 3D scene modeling, is specifically used for:

[0029] Based on semantic association, two-dimensional vector data and three-dimensional models corresponding to the same geographic entity are associated and modeled.

[0030] By using shared semantic identifiers, two-dimensional vector scene elements corresponding to the same geographic entity are associated and bound with three-dimensional model scene elements, thus completing the collaborative modeling of two-dimensional and three-dimensional scenes.

[0031] Preferably, when the 2D / 3D scene construction module dynamically adjusts the scene rendering accuracy based on the resource occupancy status fed back by the edge scene analysis and collaboration module, it is specifically used for:

[0032] The resource occupancy status of the edge scene analysis and collaboration module is collected. The resource occupancy status refers to the usage of computing resources, storage resources, and network resources of the edge device by the edge scene analysis and collaboration module during operation.

[0033] Adjust scene rendering precision based on resource usage status.

[0034] Preferably, the edge scene analysis collaboration module, when constructing a lightweight edge analysis deployment system, is specifically used for:

[0035] Based on the geographical scope and data processing requirements of the 2D / 3D twin scene, the edge analysis area is divided;

[0036] Deploy edge analytics nodes in each edge analytics region to form a distributed edge analytics network;

[0037] The analysis tasks of edge analysis nodes are broken down, enabling each edge analysis node to handle real-time analysis of corresponding 2D and 3D twin scene data, thus building a lightweight edge analysis deployment system.

[0038] Collect the resource usage status of each edge analysis node and push the resource usage status to the 2D / 3D scene construction module.

[0039] Preferably, when the edge scene analysis collaboration module performs the semantic association link between the edge analysis results and the knowledge graph, it is specifically used for:

[0040] Geographic identifiers and change characteristics are extracted from real-time analysis results, and the geographic identifiers uniquely correspond to the semantic identifiers of knowledge graph entities;

[0041] Based on geographic identifiers, the real-time analysis results are semantically matched and associated with the corresponding geographic entities and semantic association information in the geographic knowledge graph;

[0042] The system integrates the change characteristics of real-time analysis results with the geographic entity attribute information and semantic relationships matched in the knowledge graph to generate integrated analysis data, which is then pushed to the semantic module of the knowledge graph for updates.

[0043] Preferably, when the intelligent display interaction module receives and integrates the fusion analysis data pushed by the display edge scene analysis collaboration module, it is specifically used for:

[0044] Based on a geographic knowledge graph system, it provides an associated display mode, which includes associated link display, hierarchical display and thematic aggregation display;

[0045] It receives fusion analysis data pushed by the edge scene analysis collaboration module, and integrates and displays the change feature information and corresponding semantic attribute information in the fusion analysis data. The fusion display includes highlighting abnormal areas, marking change areas, and overlaying related entity information and historical related data at the marked positions.

[0046] The semantic attribute information is the geographic entity attribute information corresponding to the marked location, obtained from the geographic knowledge graph system.

[0047] Preferably, when the intelligent display and interaction module responds to a user's geographic semantic query request and displays the results in a 2D / 3D twin scene based on the query results returned by the knowledge graph semantic module, it is specifically used for:

[0048] In response to a user's geographic semantic query request, the system sends the geographic semantic query request to the knowledge graph semantic module and receives the query results, related entity information, and historical related data returned from the knowledge graph semantic module.

[0049] Locate and highlight the scene elements corresponding to the query results in the 2D / 3D twin scene, and simultaneously display the related entity information and historical related data in the interactive panel.

[0050] The present invention has the following beneficial effects:

[0051] 1. This invention constructs a collaborative architecture of four modules: 2D / 3D scene construction, knowledge graph semantics, edge scene analysis collaboration, and intelligent display interaction. It establishes a mapping mechanism between geographic data and knowledge graph entities, enabling collaborative modeling and seamless switching of 2D / 3D scenes. Furthermore, it dynamically adjusts scene rendering accuracy based on edge analysis resource status, achieving semantically integrated construction and adaptive rendering of 2D / 3D geographic scenes. Compared with existing technologies, this invention can significantly improve the relevance, semantic level, and rendering adaptability of 2D / 3D scenes. Therefore, it can solve the technical problems in existing technologies, such as fragmented 2D / 3D geographic scenes, lack of semantic identifiers for scene elements leading to low interaction accuracy, and mismatch between rendering accuracy and resource consumption, which in turn affects system operating efficiency.

[0052] 2. This invention constructs a geographic knowledge graph system by integrating government geographic data and 2D / 3D scene-related data through a knowledge graph semantic module. This enables dynamic incremental updates of the graph and semantic queries and related displays based on semantic relationships. Simultaneously, it establishes a bidirectional data interaction link between the knowledge graph and the 2D / 3D scene construction module and the edge scene analysis collaboration module. Compared with existing technologies, this significantly improves the relevance, real-time performance, and scene-related linkage of the geographic knowledge graph. Therefore, it solves the technical problems in existing technologies where geographic knowledge graphs are mostly general-purpose constructions, unable to adapt to specific 2D / 3D scene requirements, and suffer from insufficient semantic support due to lagging graph updates and low matching degree between query results and scene elements.

[0053] 3. This invention constructs a lightweight edge analysis deployment system, divides edge analysis regions, deploys distributed edge analysis nodes to decompose analysis tasks, establishes semantic association links between edge analysis results and knowledge graphs, and combines intelligent display and interaction modules to achieve multi-mode semantic association display and fusion presentation of edge analysis results and semantic information. Compared with existing technologies, it can effectively improve the real-time analysis efficiency of 2D and 3D twin scene data, the semantic interpretation capability of analysis results, and the intuitiveness of display and interaction. Therefore, it can solve the technical problems in existing technologies, such as heavy edge analysis deployment, redundant analysis tasks leading to poor real-time performance, and the separation of analysis results and semantic information, and the single display and interaction mode leading to inaccurate user perception of scene changes and related information. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the overall system architecture of the present invention;

[0055] Figure 2 This is a schematic diagram of the knowledge graph semantic module of the present invention;

[0056] Figure 3 This is a schematic diagram of the edge scene analysis and collaboration module of the present invention. Detailed Implementation

[0057] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0058] Example 1, please refer to Figure 1 and Figure 2 As shown: A digital twin-based integrated 2D / 3D geographic scene intelligent display system, with a knowledge graph semantic module, is used to construct a geographic knowledge graph system, perform semantic queries and related displays, and dynamically and incrementally update the geographic knowledge graph system; the geographic knowledge graph system includes knowledge graph entities identified by semantic tags and semantic relationships between geographic entities;

[0059] In constructing a geographic knowledge graph system, the semantic module of the knowledge graph is specifically used for:

[0060] Collect geographic data, including terrain data, image data, and vector data used to construct 2D and 3D twin scenes;

[0061] Geographic entities and their attribute information are extracted from geographic data. A geographic entity is a data unit that expresses a geographic phenomenon. Specifically, a geographic entity is a data unit that expresses a geographic phenomenon that is formed naturally, artificially constructed, or administratively managed.

[0062] The geographic entity attribute information is represented as a set of feature data of the geographic entity, specifically divided into general attribute information and classification attribute information;

[0063] The extracted geographic entities are classified, semantic relationships between different geographic entities are established, and a unique semantic identifier is assigned to each geographic entity.

[0064] Based on the classified geographic entities, geographic entity attribute information, and semantic relationships, a geographic knowledge graph system is constructed.

[0065] Knowledge graph entities correspond one-to-one with geographic entities and semantic identifiers.

[0066] The semantic module of the knowledge graph is specifically used for: performing semantic queries and displaying related information.

[0067] Based on the semantic relationships in the geographic knowledge graph system, the geographic entities associated with the semantic attributes of scene elements are queried. The scene elements are two-dimensional vector primitives or three-dimensional models that visualize geographic entities in two-dimensional and three-dimensional twin scenes.

[0068] The location and attribute tagging information of the scene elements corresponding to the associated geographic entities in the associated query results will be pushed to the intelligent display and interaction module.

[0069] When the intelligent display interaction module triggers the selection operation of the scene element corresponding to the geographic entity, it extracts the related entity information and historical related data of the geographic entity.

[0070] Related entity information refers to the attribute information of other geographic entities that have a semantic relationship with the selected geographic entity, and historical association data refers to the record data of the historical association relationship between the selected geographic entity and related geographic entities;

[0071] The extracted related entity information and historical related data are pushed to the intelligent display and interaction module.

[0072] The semantic module of the knowledge graph is specifically used for dynamically and incrementally updating the geographic knowledge graph system:

[0073] Real-time monitoring and updating of data, including fusion analysis data pushed by the edge scene analysis collaboration module and updated geographic data;

[0074] The monitored updated data is used to identify relationships, determine newly added geographic entities, newly added geographic entity attribute information and newly added semantic relationships, or identify changes in existing geographic entity attribute information and semantic relationships.

[0075] The identified new or changed content is reviewed, and once approved, it is added to the geographic knowledge graph system to complete the dynamic incremental update of the graph.

[0076] Furthermore, the specific implementation of the semantic module of the knowledge graph is as follows: In the data acquisition stage, geographic data is first collected, including terrain data, image data and vector data used to construct two-dimensional and three-dimensional twin scenes. The collected geographic data is standardized and preprocessed, including data format unification, redundant information removal and data validity verification, laying the foundation for subsequent processing.

[0077] Geographic data is acquired through multi-source data acquisition channels to ensure the comprehensiveness and reliability of the data source. The raw terrain data comes from satellite remote sensing observation systems, aerial photogrammetry data platforms, etc. After being acquired through data subscription or batch retrieval, preprocessing operations are performed, including data format conversion and redundant noise removal, to finally form standardized terrain data.

[0078] Image data comes from high-resolution satellite image databases, drone aerial photography systems, etc. It is received in real time or extracted in batches through standardized data interfaces. After extraction, routine processing such as color correction and resolution unification is required to ensure the consistency of image data.

[0079] Vector data comes from the government geographic information public service platform and the basic surveying and mapping results database. The extraction process must follow the government data sharing standards. After extraction, preprocessing work such as coordinate system I and topological relationship checks must be completed.

[0080] In the stage of extracting geographic entities and attribute information, geographic entities and geographic entity attribute information are extracted from the preprocessed geographic data. A geographic entity is a data unit that expresses geographic phenomena. Specifically, a geographic entity is a data unit that expresses geographic phenomena formed naturally, artificially constructed, or administratively managed.

[0081] Geographic entity attribute information is represented as a set of feature data of geographic entities, specifically divided into general attribute information and classification attribute information;

[0082] General attribute information includes entity identification feature data, spatial location feature data, data management feature data, and association relationship feature data;

[0083] The specific entity identifier feature data consists of the entity identifier's name and abbreviation;

[0084] Spatial positioning feature data specifically includes geometric type, coordinate range, projected coordinate system, and elevation parameters; data management feature data specifically includes collection time, update time, data source, and responsible entity; and correlation feature data specifically includes spatial relationships with other geographic entities.

[0085] The classification attribute information includes natural feature data, artificial construction feature data, and administrative management feature data; the natural feature data specifically includes landform type, altitude and slope, watershed area and runoff, vegetation type and coverage, rock type and geological age, and climate parameters;

[0086] The specific data on artificial construction features include building structure and use, road grade and pavement material, reservoir capacity and flood control standards, hub throughput and operating time, and industrial and mining capacity and emission indicators; the specific data on administrative management features include administrative level and jurisdiction, grid code and grid member information, property rights and usage period, population and GDP.

[0087] Subsequently, semantic relationships between geographic entities are established. Specifically, semantic matching is performed based on the classification attribute information of geographic entities to establish type and functional relationships, or administrative relationships are established by sorting out administrative affiliation and jurisdictional hierarchy based on the management attribute information of geographic entities, or time-series relationships are established by mining the interaction records of geographic entities in the time dimension, such as time-series data on water supply relationships between water conservancy facilities and watersheds, and traffic relationship data between transportation networks and settlements.

[0088] The specific implementation of semantic matching based on classification attribute information is as follows: First, extract the core features of the classification attributes of each geographic entity, such as landform type and vegetation coverage in natural feature data, building use and road grade in artificial construction feature data, and administrative level in administrative management feature data, and construct a semantic matching rule base. The rule base includes rules for associating entities of the same type, rules for associating entities with complementary functions, and rules for causal association. Rules for associating entities of the same type include roads and bridges belonging to the transportation facility category; rules for associating entities with complementary functions include reservoirs and irrigation canals; and rules for causal association include mining areas and surrounding areas with changes in vegetation cover. By matching the classification attribute features of geographic entities with the rule base, the semantic association type between entities is determined, and then type association or functional association is established.

[0089] The establishment of administrative associations is based on administrative management characteristic data. It extracts attribute information such as administrative level, jurisdiction, and power nature of geographic entities, clarifies the subordinate relationship of administrative entities at different levels, such as provincial-level administrative regions, municipal-level administrative regions, and county-level administrative regions, as well as the jurisdictional relationship between administrative entities and natural and artificial geographic entities within their jurisdiction, such as a municipal government and parks, roads, and industrial and mining enterprises within its jurisdiction. These relationships are then transformed into standardized semantic association expressions and incorporated into the semantic association relationship system.

[0090] The establishment of temporal correlations relies on historical correlation data of geographic entities. By filtering the interaction behavior data between geographic entities in different time periods through timestamps, such as the correlation data between reservoir water level changes and downstream river runoff during the flood season, and the correlation data between road traffic volume and pedestrian flow in surrounding business districts during morning and evening peak hours, the temporal correlation between data is analyzed to determine the dynamic correlation between entities over time, and time attribute labels are added to these correlations to realize the construction of temporal correlations.

[0091] After establishing various semantic relationships, and combining the geographic entity classification results and attribute information mentioned above, a unique semantic identifier is assigned to each geographic entity to ensure that the semantic identifier corresponds one-to-one with the geographic entity. Finally, the classified geographic entities, complete geographic entity attribute information, and multi-dimensional semantic relationships are integrated to form a complete and clearly related geographic knowledge graph system, providing basic support for subsequent semantic query, related display, and dynamic incremental updates.

[0092] When performing semantic queries and related displays, the knowledge graph semantic module receives geographic semantic query requests from the intelligent display interaction module, parses the semantic keywords in the request, and matches the corresponding knowledge graph entities in the geographic knowledge graph system. It then performs a traversal query based on semantic relationships to obtain related geographic entity information. The module pushes the location and attribute tagging information of the scene elements corresponding to the related geographic entities to the intelligent display interaction module. When the intelligent display interaction module triggers a scene element selection operation, it matches the corresponding knowledge graph entity through the semantic identifier associated with the scene element, extracts the entity's related entity information and historical related data, and pushes this information to the intelligent display interaction module for display.

[0093] When dynamically and incrementally updating the geographic knowledge graph system, the updated data is monitored in real time. The updated data includes fusion analysis data pushed by the edge scene analysis collaboration module and incremental update data from multi-source geographic data acquisition channels. By comparing with the existing graph content, new geographic entities, new attribute information, new semantic relationships, or changes in the attributes and relationships of existing geographic entities are identified in the updated data. The identified new or changed content is reviewed, and after approval, it is added to or updated into the geographic knowledge graph system to complete the dynamic incremental update and maintain the real-time performance and accuracy of the graph.

[0094] Example 2, please refer to Figure 1 As shown: A digital twin-based integrated 2D / 3D geographic scene intelligent display system.

[0095] Furthermore, the 2D modeling stage employs conventional vector drawing tools to construct the 2D outline of geographic entities based on preprocessed vector data. During the drawing process, the acquired semantic identifiers are synchronously associated with the corresponding 2D vector objects, ensuring that each 2D vector object carries a unique semantic identifier. In the 3D modeling stage, for the same geographic object, a 3D model is constructed using 3D geometric modeling techniques based on the 2D vector outline data. Basic details such as texture and material of the model are supplemented. During the modeling process, it is strictly ensured that the semantic identifier of the 3D model is completely consistent with the semantic identifier of the corresponding 2D vector object, achieving a unique semantic identifier binding between 2D and 3D data.

[0096] During the transition between 2D and 3D scenes, the semantic identifiers and semantic relationships of scene elements remain unchanged, so that the scene elements after the transition can still be associated with the corresponding knowledge graph entities and semantic attributes through semantic identifiers.

[0097] Scene switching is primarily handled by a standard switching control module. Before switching, this module first reads the current scene's view status information, including core parameters such as viewpoint position, zoom level, and the range of currently displayed scene elements. Then, based on an association mapping table, it uses the semantic identifiers of the scene elements displayed in the current scene as query keywords to quickly match the corresponding scene elements in the other dimension, ensuring consistency in the displayed geographic object range before and after the switch. After matching, the standard scene rendering engine is invoked to complete the view switch according to the preset transition effects. During the switch, the semantic identifiers and semantic relationships of the scene elements are strictly maintained unchanged. After the switch, scene elements can directly retrieve the corresponding knowledge graph entity attributes and relationship data from the geographic knowledge graph system through their semantic identifiers, achieving continuity of semantic information before and after the switch and ensuring a consistent user experience.

[0098] Resource occupancy status is collected via the resource monitoring unit of the edge devices. A conventional resource monitoring agent is deployed at each edge analysis node. This agent can collect core resource metrics of the edge devices in real time, including CPU utilization, memory utilization, storage space utilization, and network bandwidth utilization. The collected data is pushed to the resource analysis unit of the 2D / 3D scene construction module in real time via conventional lightweight data transmission methods. The push cycle can be flexibly configured according to the real-time requirements of scene construction. To ensure the accuracy of the collected data, the resource analysis unit needs to verify the validity of the pushed resource occupancy status data. Conventional data verification methods are used to remove abnormal fluctuations caused by momentary device failures or network fluctuations. For missing data, conventional interpolation methods are used to supplement it, ensuring the integrity and reliability of the resource data and providing an accurate basis for subsequent accuracy adjustments.

[0099] Adjusting rendering accuracy requires establishing a dynamic adjustment rule system. First, preset resource usage threshold ranges using conventional parameter configuration methods. Divide core indicators such as CPU utilization and memory utilization into three ranges: low load, medium load, and high load. The low load range for CPU utilization is no more than 30%, the medium load range is 30% to 70%, and the high load range is more than 70%. The low load range for memory utilization is no more than 40%, the medium load range is 40% to 80%, and the high load range is more than 80%.

[0100] When resource usage is in a low-load range, scene rendering accuracy is improved. This is achieved through conventional methods such as increasing the geometric details of 3D models, enabling high-definition texture display, and increasing the display resolution of image data. Simultaneously, it supports rendering detailed effects such as shadows and lighting to enhance the visual quality of the scene. When resource usage is in a medium-load range, the current rendering accuracy is maintained to ensure a balance between scene visualization and system efficiency. When resource usage is in a high-load range, scene rendering accuracy is reduced. This is achieved by simplifying the geometric structure of 3D models, using compressed texture maps, and reducing the sampling rate of image data to reduce rendering computation. At the same time, priority is given to preserving the rendering details of core geographic entities, including major roads and key buildings, to avoid the loss of critical geographic information due to reduced accuracy. Changes in rendering accuracy are implemented through the scene rendering engine's parameter configuration interface. After adjustments are made, the resource analysis unit receives real-time feedback from the rendering engine regarding the adjustment effect.

[0101] Example 3, please refer to Figure 1 and Figure 3 As shown: A digital twin-based integrated 2D / 3D geographic scene intelligent display system.

[0102] Furthermore, the specific implementation process and core logic of the edge scene analysis collaboration module are as follows: In the construction phase of the lightweight edge analysis deployment system, the first step is to formulate division rules based on the geographical scope of the two-dimensional and three-dimensional twin scenes and the data processing requirements. The geographical grid division method adapted to the geographical scene is used to divide the non-overlapping areas. Then, the boundaries are calibrated in combination with administrative boundaries and natural geographical barriers to ensure that the workload of each area is balanced.

[0103] After completing the regional division, lightweight edge hardware nodes are selected according to the regional task load. The selection criteria are based on the minimum configuration that meets the real-time analysis requirements of the region, adapts to the real-time analysis task requirements of corresponding 2D and 3D twin scene data, and prioritizes deployment close to the data acquisition terminal to ensure real-time performance. Each node forms a distributed edge analysis network through a low-latency, high-reliability communication protocol. At the same time, a regional coordination node is set up to be responsible for task scheduling and status monitoring. The coordination nodes interact and collaborate through the backbone link to form a distributed deployment architecture.

[0104] After node deployment, following the principles of region exclusivity and task minimization, tasks are broken down according to data type and analysis function. Data types include terrain data, image data, and vector data. Analysis functions include data preprocessing, feature extraction, and change detection. Dedicated task lists are assigned, and interference is avoided through a dual isolation mechanism based on task type and regional scope. After task decomposition, tasks are distributed through the scheduling platform. The scheduling platform interacts with the regional coordination node in real time, verifies node load, and completes task implementation, ensuring that each edge analysis node processes the corresponding 2D / 3D twin scene data in real time. During this process, the resource occupancy status of each edge analysis node is collected through the node's built-in monitoring agent, covering computing, storage, and network resources. The collection cycle is adaptively adjusted according to the peak data processing time. After standardization and anomaly removal, the data is organized into a unified format and then pushed to the 2D / 3D scene construction module in real time through an encrypted link, while a backup is stored on the coordination node.

[0105] During the stage of establishing the semantic association link between edge analysis results and knowledge graph, after the real-time analysis results are generated, the geographic identifiers and change feature information are extracted by the feature extraction algorithm adapted to the geographic scene. The geographic identifiers adopt the same coding standard as the semantic module of the knowledge graph. Consistency is ensured through pre-deployment negotiation and periodic verification after deployment, so as to ensure that the geographic identifiers and the semantic identifiers of knowledge graph entities correspond uniquely.

[0106] Semantic matching is initiated based on the extracted geographic identifiers. The semantic mapping table and spatial location information are used to assist in the verification to ensure the accuracy of the matching. If the matching fails, the fault tolerance mechanism is triggered to perform secondary matching and anomaly reporting to ensure the stability of the association link. The real-time analysis results are then linked with the corresponding geographic entities and semantic association information in the geographic knowledge graph through the matching.

[0107] Finally, according to the preset rules of attribute correlation priority and change timeliness, the change feature information in the real-time analysis results is structurally fused with the geographic entity attribute information and semantic relationship matched in the knowledge graph. After integrity verification, the data is packaged into a fused analysis data in a format compatible with the knowledge graph semantic module, and then pushed to the knowledge graph semantic module through a dedicated interface to provide accurate support for the dynamic incremental update of the graph.

[0108] Example 4, please refer to Figure 1 As shown: A digital twin-based integrated 2D / 3D geographic scene intelligent display system.

[0109] Furthermore, the specific implementation process and core logic of the intelligent display and interaction module are as follows: Multi-mode semantic association display is based on the geographic entity and relationship data provided by the knowledge graph semantic module to build a dedicated display engine. The engine has built-in mode adaptation rules, which can automatically switch or manually select the association display mode according to user operation needs and application scenarios.

[0110] The association link display uses differentiated visual lines to outline the semantic association path between geographical entities. The line type and color are designed to distinguish the association type as subordinate, adjacent, or inclusive. At the same time, the core attribute name and function type of the entity are marked on the link node to intuitively present the semantic association logic between entities.

[0111] The hierarchical display constructs a tree-like display structure based on the inherent characteristics of geographic entities, such as administrative and functional levels. The tree structure is precisely associated with the spatial location of the 2D and 3D twin scenes, allowing users to expand and collapse lower-level entities by clicking on nodes, and quickly locate geographic entities at different levels and their corresponding related information.

[0112] The thematic aggregation display categorizes and aggregates geographic entities according to preset geographic themes such as transportation facilities, public service facilities, and physical geography. It also supports users to customize thematic classification dimensions based on geographic entity attributes. Each theme is configured with an independent display entry and label style. Users can quickly switch between themes through the theme filtering menu to obtain complete semantic association data and spatial distribution of the target category entities.

[0113] In the process of integrating and displaying edge analysis results and semantic information, the integrated analysis data pushed by the edge scene analysis collaboration module is first parsed to extract the geographic identifiers, change feature information and related semantic attribute information. Based on the geographic identifiers, the corresponding spatial locations in the two-dimensional and three-dimensional twin scenes are matched to construct the display association mapping.

[0114] For abnormal areas, a preset differentiated highlighting style system is adopted, matching different color saturation and transparency highlighting effects according to the severity of the abnormality. At the same time, a floating information window is bound to the marked area to display core information such as abnormality type, change range, and occurrence time in real time. For changed areas, a dynamic icon encoding mechanism is adopted, with different change types corresponding to exclusive dynamic icons. The icon positions are accurately anchored to the spatial coordinates of geographic objects to ensure no deviation from the entity location. The superimposed related semantic information is organized according to the principle of prioritizing core attributes and hierarchical related information, and presented in the form of a concise text pop-up. The pop-up supports automatic collapse and manual expansion. Clicking the pop-up will jump to the full semantic information display page.

[0115] When responding to a geographic semantic query, the system first receives geographic semantic query requests initiated by users through multiple modes such as voice, text, and gestures. It extracts the core semantic keywords from the request and encapsulates them in a standardized format supported by the knowledge graph semantic module. This encapsulates the requests and sends them to the knowledge graph semantic module via a dedicated communication interface. After receiving the query results, related entity information, and historical related data returned by the knowledge graph semantic module, the system extracts the geographic identifiers, spatial coordinates, and related data. This triggers the positioning and jump function of the 2D / 3D twin scene, automatically switching the scene view to the location of the target entity and highlighting the corresponding scene elements with a prominent style. Simultaneously, a structured information display panel is generated in the scene sidebar. The panel categorizes and displays information according to the core attribute area, the related entity list area, and the historical related data area. Users can quickly jump to the spatial location of the corresponding entity by clicking on the related entity entries listed in the panel, enabling batch viewing of related entities and scene navigation.

[0116] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A two-dimensional and three-dimensional integrated intelligent display system for geographic scenes based on digital twins, characterized in that: The system includes: The knowledge graph semantic module is used to construct a geographic knowledge graph system, perform semantic queries and related displays, and dynamically and incrementally update the geographic knowledge graph system; the geographic knowledge graph system includes knowledge graph entities identified by semantic tags and semantic relationships between geographic entities; The 2D / 3D scene construction module performs collaborative modeling of 2D / 3D scenes based on semantic relationships, and dynamically adjusts the scene rendering accuracy based on the resource usage status fed back by the edge scene analysis and collaboration module. The edge scene analysis collaboration module is used to build a lightweight edge analysis deployment system, perform real-time analysis of 2D and 3D twin scene data, establish semantic association links between edge analysis results and knowledge graph, generate fusion analysis data and push it to the knowledge graph semantic module, and collect its own resource occupancy status and push it to the 2D and 3D scene construction module. The intelligent display and interaction module is used to receive and integrate the fusion analysis data pushed by the display edge scene analysis and collaboration module, respond to the user's geographic semantic query request, and display the related data in the two-dimensional twin scene based on the query results returned by the knowledge graph semantic module. The 2D / 3D scene construction module is specifically used for: During collaborative 2D / 3D scene modeling, Based on semantic association, two-dimensional vector data and three-dimensional models corresponding to the same geographic entity are associated and modeled. By using shared semantic identifiers, two-dimensional vector scene elements corresponding to the same geographic entity are associated and bound with three-dimensional model scene elements to complete collaborative modeling of two-dimensional and three-dimensional scenes. When the edge scene analysis collaboration module establishes a semantic association link between the edge analysis results and the knowledge graph, it is specifically used for: Geographic identifiers and change characteristics are extracted from real-time analysis results, and the geographic identifiers uniquely correspond to the semantic identifiers of knowledge graph entities; Based on geographic identifiers, the real-time analysis results are semantically matched and associated with the corresponding geographic entities and semantic association information in the geographic knowledge graph; The system integrates the change characteristics of real-time analysis results with the geographic entity attribute information and semantic relationships matched in the knowledge graph to generate integrated analysis data, which is then pushed to the semantic module of the knowledge graph for updates.

2. The intelligent display system for integrated 2D and 3D geographic scenes based on digital twins according to claim 1, characterized in that, The semantic module of the knowledge graph is specifically used in constructing a geographic knowledge graph system for: Collect geographic data, including terrain data, image data, and vector data for constructing 2D and 3D twin scenes; Geographic entities and their attribute information are extracted from geographic data. A geographic entity is a data unit that expresses a geographic phenomenon. Specifically, a geographic entity is a data unit that expresses a geographic phenomenon that is formed naturally, artificially constructed, or administratively managed. The geographic entity attribute information is represented as a set of feature data of the geographic entity, specifically divided into general attribute information and classification attribute information; The extracted geographic entities are classified, semantic relationships between different geographic entities are established, and a unique semantic identifier is assigned to each geographic entity. Based on the classified geographic entities, geographic entity attribute information, and semantic relationships, a geographic knowledge graph system is constructed. The knowledge graph entities correspond one-to-one with geographic entities and semantic identifiers.

3. The intelligent display system for integrated 2D and 3D geographic scenes based on digital twins according to claim 1, characterized in that, The semantic module of the knowledge graph is specifically used for: performing semantic queries and displaying related information. Based on the semantic relationships in the geographic knowledge graph system, the geographic entities associated with the semantic attributes of scene elements are queried. The scene elements are two-dimensional vector primitives or three-dimensional models that visualize geographic entities in a two-dimensional or three-dimensional twin scene. The location and attribute tagging information of the scene elements corresponding to the associated geographic entities in the associated query results will be pushed to the intelligent display and interaction module. When the intelligent display interaction module triggers the selection operation of the scene element corresponding to the geographic entity, it extracts the related entity information and historical related data of the geographic entity. The associated entity information is the attribute information of other geographic entities that have a semantic relationship with the selected geographic entity, and the historical association data is the record data of the historical association relationship between the selected geographic entity and the associated geographic entities. The extracted related entity information and historical related data are pushed to the intelligent display and interaction module.

4. The intelligent display system for integrated 2D and 3D geographic scenes based on digital twins according to claim 1, characterized in that, The semantic module of the knowledge graph is specifically used for dynamically and incrementally updating the geographic knowledge graph system as follows: Real-time monitoring of updated data, including fusion analysis data pushed by the edge scene analysis collaboration module and updated geographic data; The monitored updated data is used to identify relationships, determine newly added geographic entities, newly added geographic entity attribute information and newly added semantic relationships, or identify changes in existing geographic entity attribute information and semantic relationships. The identified new or changed content is reviewed, and once approved, it is added to the geographic knowledge graph system to complete the dynamic incremental update of the graph.

5. The intelligent display system for integrated 2D and 3D geographic scenes based on digital twins according to claim 1, characterized in that, The 2D / 3D scene construction module, when dynamically adjusting scene rendering precision based on resource usage status feedback from the edge scene analysis and collaboration module, is specifically used for: The resource occupancy status of the edge scene analysis and collaboration module is collected. The resource occupancy status refers to the usage of computing resources, storage resources, and network resources of the edge device by the edge scene analysis and collaboration module during operation. Adjust scene rendering precision based on resource usage status.

6. The intelligent display system for integrated 2D and 3D geographic scenes based on digital twins according to claim 1, characterized in that, The edge scene analysis collaboration module is specifically used in building a lightweight edge analysis deployment system for: Based on the geographical scope and data processing requirements of the 2D / 3D twin scene, the edge analysis area is divided; Deploy edge analytics nodes in each edge analytics region to form a distributed edge analytics network; The analysis tasks of edge analysis nodes are broken down, enabling each edge analysis node to handle real-time analysis of corresponding 2D and 3D twin scene data, thus building a lightweight edge analysis deployment system. Collect the resource usage status of each edge analysis node and push the resource usage status to the 2D / 3D scene construction module.

7. The intelligent display system for integrated 2D and 3D geographic scenes based on digital twins according to claim 1, characterized in that, When the intelligent display interaction module receives and integrates the fusion analysis data pushed by the display edge scene analysis collaboration module, it is specifically used for: Based on a geographic knowledge graph system, it provides an associated display mode, which includes associated link display, hierarchical display and thematic aggregation display; It receives fusion analysis data pushed by the edge scene analysis collaboration module, and integrates and displays the change feature information and corresponding semantic attribute information in the fusion analysis data. The fusion display includes highlighting abnormal areas, marking change areas, and overlaying related entity information and historical related data at the marked positions. The semantic attribute information is the geographic entity attribute information corresponding to the marked location, obtained from the geographic knowledge graph system.

8. The intelligent display system for integrated 2D and 3D geographic scenes based on digital twins according to claim 1, characterized in that, When the intelligent display and interaction module responds to a user's geographic semantic query request and displays the results in a 2D / 3D twin scene based on the query results returned by the knowledge graph semantic module, it is specifically used for: In response to a user's geographic semantic query request, the system sends the geographic semantic query request to the knowledge graph semantic module and receives the query results, related entity information, and historical related data returned from the knowledge graph semantic module. Locate and highlight the scene elements corresponding to the query results in the 2D / 3D twin scene, and simultaneously display the related entity information and historical related data in the interactive panel.

Images