A knowledge-guided dynamic expression method for urban physical examination scenarios

Abstract

The present invention provides a knowledge-guided dynamic expression method for urban physical examination scenarios. The method obtains business demand data of a target city based on the target city's spatiotemporal data, and constructs a scenario knowledge graph based on the target city's spatiotemporal data and the business demand data; the target city's spatiotemporal data and the scenario knowledge graph are parsed to obtain parsing results, and the parsing results are reorganized by combining semantic constraints with graphic element combinations, and the scenario narrative logic is used to drive the scene generation and update to obtain an urban physical examination scenario; the semantics of objects in the urban physical examination scenario are enhanced in a dynamic and static manner to obtain an enhanced urban physical examination scenario, and the enhanced urban physical examination scenario is presented with data priority scheduling to obtain a dynamic expression result of the urban physical examination scenario; the problems in the existing technology of disconnection between business and data, insufficient adaptability of dynamic scene expression, and relatively single visualization presentation are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban physical examination scene expression, and in particular to a knowledge-guided urban physical examination scene dynamic expression method. Background Art

[0002] The core task of urban health checks is to quantify, calculate, analyze, and evaluate various data such as urban spatial structure, population distribution, resource supply and demand, and environmental quality, in order to reveal the health status and development trends of urban systems and serve urban planning and management. In this process, the dynamic changes in urban vital signs are presented through urban health check scenarios: urban digital application contexts composed of specific geographical and humanistic attributes, which collaborate with the knowledge, vision, and operations of the health check process. By establishing a two-way coupling between decision makers' understanding of urban health check scenarios and the refined expression of urban vital signs, urban health check scenarios transform complex spatiotemporal data into intuitive and easy-to-understand dynamic visualization maps, interactive infographics, concise and accurate text, and other presentation forms, providing decision support for the formulation of urban planning and management measures.

[0003] However, most existing urban scenario models still adopt a data-centric approach, with insufficient knowledge linkage. This makes it difficult to fully connect decision makers with the necessary urban health scenario knowledge, resulting in complex spatiotemporal data for urban health assessments. Furthermore, as urban physical signs continue to change, dynamic adjustments to urban health scenario elements are required, such as the events, processes, states, and related objects and people within the scenario. However, current visualization of these dynamic changes is ineffective, resulting in poor spatiotemporal data presentation.

[0004] In fact, in urban digital governance, urban health examination scenarios have both dynamic and static characteristics. Static is reflected in some relatively stable basic elements in the scene, such as the city's geographical location, topography, and long-term functional zoning; while dynamic is manifested in the continuous changes in urban signs and the dynamic expression of scene elements caused by this. From a technical perspective, knowledge graphs can effectively integrate the business needs and spatiotemporal data of urban health examination scenarios to achieve business and data association; spatiotemporal narratives give time and logical clues to the semantic information of urban health examination scenarios, connecting isolated urban phenomena into coherent stories to enhance readability and understanding; maps and virtual geographic environments provide intuitive visual display methods for geographic space scene information in urban health examination scenarios, making complex information easier to accept. The combined use of these technologies provides strong support for the dynamic expression of urban health examination scenarios. However, in actual application, there are many difficulties, such as:

[0005] (1) Disconnection between data and business needs: During the design and implementation process, existing knowledge graph methods focus more on the association with data and lack scenario knowledge for business collaboration, resulting in a disconnect between the expression results and business needs. Due to the lack of scenario knowledge, the potential semantic relationships between many urban physical examination data and knowledge have not been effectively mined and extracted, affecting the accuracy and completeness of scenario expression.

[0006] (2) Insufficient adaptability of dynamic expression: Existing scene expression methods often show obvious lack of adaptability when facing these diverse scene changes. They are usually based on fixed arrangement patterns and data-driven frameworks, lacking sufficient flexibility and scalability. Secondly, traditional urban physical examination reports are mostly static data lists, lacking plot and coherence. Even in the narrative of certain urban phenomena, the time and space dimensions are often separated from each other, making it difficult to clearly present the temporal and spatial causal relationship, which is not conducive to in-depth analysis of urban problems.

[0007] (3) The visualization presentation is relatively simple: the map can present basic urban health information, but the amount of information carried by the two-dimensional map is limited, and it is difficult to support the dynamic visualization of the urban health scene; in addition, the human eye has a certain dependence on perceptual saliency when extracting information, and dynamic visual phenomena such as flickering, jumping and changing can attract more attention than static presentation. Therefore, the traditional static visualization method has obvious limitations in the effect of information transmission. Summary of the Invention

[0008] The present invention provides a knowledge-guided method for dynamically expressing urban physical examination scenarios, which aims to solve the problems in the existing technology of disconnection between business and data, insufficient adaptability of dynamic scene expression, and relatively simple visualization presentation.

[0009] To achieve the above objectives, the present invention provides a knowledge-guided method for dynamically expressing urban physical examination scenarios, comprising:

[0010] Step 1: Obtain the target city's business demand data based on the target city's spatiotemporal data, and construct a scenario knowledge graph based on the target city's spatiotemporal data and business demand data. The business demand data includes knowledge-based assessment data, spatiotemporal analysis data, and diversified evidence data.

[0011] Step 2: Analyze the target city's spatiotemporal data and scenario knowledge graph to obtain analysis results. Combine semantic constraints with graphic element combinations to guide the analysis results for reorganization. Drive scenario generation and update with the help of scenario narrative logic to obtain the city physical examination scenario.

[0012] Step 3: Enhance the semantics of objects in the urban physical examination scene by combining static and dynamic methods to obtain an enhanced urban physical examination scene, and perform data-priority scheduling and presentation on the enhanced urban physical examination scene to obtain a dynamic expression result of the urban physical examination scene.

[0013] More specifically, step 1 includes:

[0014] Step 11: Acquire the business demand data of the target city based on the spatiotemporal data of the target city;

[0015] Step 12: Build a hypergraph-based architecture layer based on business demand data and target city spatiotemporal data;

[0016] Step 13: Entity classification and entity relationship construction are performed on the architecture layer, and knowledge fusion of different entities is performed using hypernode links and hyperedge links to obtain the initial scene knowledge graph. The scene knowledge graph includes a semantic layer, a data layer, and a scene layer.

[0017] Step 14: Use a hierarchical storage architecture to store each semantic entity in the semantic layer as a scene knowledge graph node, store the original data in the data layer, and store the data in the scene layer in the form of an attribute table. Use a unique identifier and index mechanism to associate the data in the semantic layer, data layer, and scene layer. When the spatiotemporal data of the target city changes, the initial scene knowledge graph is dynamically updated to obtain a scene knowledge graph.

[0018] To further elaborate, the formal expression of the architecture layer is:

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] in, Represents the architectural layer, Represents a semantic set, Represents the topic node, Represents the connotation node, represents a knowledge node, Represents an indicator node, Represents a data set, Indicates the working stage of the city physical examination. Indicates the data type, Indicates the source of data. Represents data attributes, Represents the relationship between data, Represents a collection of application scenarios, Indicates the type of application scenario. Represents the hierarchical structure of data in the urban physical examination process, represents a data subgraph consisting of a set of nodes and edges, Represents higher-order relations in a data hypergraph.

[0024] Furthermore, when the spatiotemporal data of the target city changes, the initial scene knowledge graph is dynamically updated to obtain a scene knowledge graph, including:

[0025] When the spatiotemporal data of the target city changes, based on the existing structure and semantic rules of the initial scene knowledge graph, incremental updates are used to adjust the entities, relationships, and attributes in the initial scene knowledge graph. After performing conflict or consistency checks on the adjusted initial scene knowledge graph, the scene knowledge graph is obtained.

[0026] When the target city's spatiotemporal data needs to be comprehensively revised, the architecture layer and the initial scenario knowledge graph are updated and stored based on the revised target city's spatiotemporal data to obtain the scenario knowledge graph.

[0027] More specifically, step 2 includes:

[0028] Step 21: parse the target city's spatiotemporal data and scene knowledge graph to obtain multiple material elements, and integrate and manage all the material elements to form a scene material library;

[0029] Step 22: Decompose the scene in the scene knowledge graph to obtain multiple basic modules. Combine semantic constraints to perform element mapping on all basic modules, all material elements in the scene material library, and semantic relationships in the scene knowledge graph to obtain multiple mapping results. All mapping results are assembled to obtain a scene model.

[0030] Step 23, dynamically evolve the scenario model through scenario narrative logic modeling to obtain the evolution result, transform the evolution result through expression processing to obtain the initial city physical examination scenario, and update the initial city physical examination scenario through the dynamic update mechanism to obtain the city physical examination scenario.

[0031] Furthermore, the initial city physical examination scenario is updated through a dynamic update mechanism to obtain a city physical examination scenario, including:

[0032] By formula Dynamically update the material primitives to obtain the updated scene material library, where: Represents the material element, Indicates the parsing method. represents the target city’s spatiotemporal data, Represents scene knowledge graph;

[0033] By formula Dynamically update the scene model to obtain an updated scene model, where: Represents the scene model, represents a composition operation, Indicates the mapped primitives;

[0034] By formula The initial city physical examination scene is updated to obtain the city physical examination scene, where: Indicates the initial city physical examination scenario, Represents visualization operations, Represents the timeline, Represents a geospatial scene.

[0035] More specifically, step 3 includes:

[0036] Step 31, enhancing the semantics of objects in the urban physical examination scene based on the combination of static and dynamic visual variables to obtain an enhanced urban physical examination scene;

[0037] Step 32: Prioritize the data in the enhanced urban physical examination scenario based on the importance and business requirements of the data in the urban physical examination scenario to obtain a priority result;

[0038] Step 33: Schedule and present the data of the enhanced urban physical examination scene according to the division results to obtain a dynamic expression result of the urban physical examination scene.

[0039] Furthermore, the formal expression of enhancing the semantics of objects in urban physical examination scenarios based on the combination of dynamic and static visual variables is as follows:

[0040] ;

[0041] in, represents a set of visual variables, represents dynamic visual variables, represents static visual variables, represents the scene features for enhanced representation, Represents the feature information of scene objects, Represents spatial location information, Indicates attribute information. Indicates related information.

[0042] Furthermore, based on the importance and business needs of the data in the urban physical examination scenario, the formal expression of the priority division of the data in the enhanced urban physical examination scenario is as follows:

[0043] ;

[0044] ;

[0045] in, Indicates high priority data, Represents the feature information of geographic scene objects, Represents the feature information of the video object, Indicates the characteristic information of dynamic symbols, Indicates low priority data, Represents the characteristic information of the icon object. Represents the feature information of a text object.

[0046] Furthermore, the formal expression for scheduling and presenting the data of the enhanced urban physical examination scenario according to the division results is:

[0047] ;

[0048] in, object feature data representing the scene, Indicates the scene scheduling and presentation method, Indicates the priority division result, Indicates the data display method. Indicates the publishing method.

[0049] The above solution of the present invention has the following beneficial effects:

[0050] The present invention obtains the business demand data of the target city based on the target city's spatiotemporal data, and constructs a scenario knowledge graph based on the target city's spatiotemporal data and the business demand data; the target city's spatiotemporal data and the scenario knowledge graph are parsed to obtain parsing results, and the parsing results are reorganized by combining semantic constraints with graphic element combinations, and the scene is generated and updated with the help of scene narrative logic to obtain a city physical examination scene; the object semantics in the city physical examination scene is enhanced in a dynamic and static manner to obtain an enhanced city physical examination scene, and the enhanced city physical examination scene is presented with data priority scheduling to obtain a dynamic expression result of the city physical examination scene; compared with the prior art, the present invention is The introduction of knowledge graph to construct scenario knowledge graph deeply associates urban spatiotemporal data and business demand data, which solves the problem of disconnection between business and data in existing technologies. The combination of semantic constraints and graphic element combinations guides the reorganization of target city spatiotemporal data and scenario knowledge graph analysis results. The scenario narrative logic is used to drive scene generation and update, which solves the problem of insufficient adaptability of dynamic scene expression. The object semantics in the urban physical examination scene is enhanced in a dynamic and static combination method to obtain an enhanced urban physical examination scene, and the enhanced urban physical examination scene is presented with data priority scheduling, which significantly improves the visualization results of the urban physical examination scene and solves the problem of relatively single visualization presentation.

[0051] Other beneficial effects of the present invention will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of a flow chart of an embodiment of the present invention;

[0053] Figure 2 This is a flow chart of a modeling method according to an embodiment of the present invention;

[0054] Figure 3 A flowchart of knowledge graph construction in an embodiment of the present invention;

[0055] Figure 4 This is a flowchart generated for the urban physical examination scenario in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] To make the technical problems, technical solutions, and advantages to be solved by the present invention more clear, the following is a detailed description with reference to the accompanying drawings and specific embodiments. It is obvious that the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0057] In the description of the present invention, it should be noted that the terms "first", "second" and "third" are only used for descriptive purposes and should not be understood as indicating or implying relative importance.

[0058] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "connected" and "connection" should be understood in a broad sense. For example, they can refer to a locking connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0059] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0060] In response to existing problems, the present invention provides a knowledge-guided method for dynamically expressing urban physical examination scenarios.

[0061] like Figure 1 、 Figure 2 As shown, an embodiment of the present invention provides a knowledge-guided method for dynamically expressing urban physical examination scenarios, including:

[0062] Step 1: Obtain the target city's business demand data based on the target city's spatiotemporal data, and construct a scenario knowledge graph based on the target city's spatiotemporal data and business demand data. The business demand data includes knowledge-based assessment data, spatiotemporal analysis data, and diversified evidence data.

[0063] Step 2: Analyze the target city's spatiotemporal data and scenario knowledge graph to obtain analysis results. Combine semantic constraints with graphic element combinations to guide the analysis results for reorganization. Drive scenario generation and update with the help of scenario narrative logic to obtain the city physical examination scenario.

[0064] Step 3: Enhance the semantics of objects in the urban physical examination scene by combining static and dynamic methods to obtain an enhanced urban physical examination scene, and perform data-priority scheduling and presentation on the enhanced urban physical examination scene to obtain a dynamic expression result of the urban physical examination scene.

[0065] Since the business process of urban physical examination is complex and diverse, covering multiple key links such as data processing, indicator calculation, evaluation and analysis, result reporting and feedback, etc., each link has significant differences in the form and content requirements of knowledge expression. In order to cope with the complex spatiotemporal data and business needs of urban physical examination, the embodiment of the present invention uses the high-order relationship expression ability of hypergraphs to directly connect multiple nodes through hyperedges to construct a scene knowledge graph, such as Figure 3 As shown, the following steps are included:

[0066] Step 11: Acquire the business demand data of the target city based on the spatiotemporal data of the target city;

[0067] Step 12: Build a hypergraph-based architecture layer based on business demand data and target city spatiotemporal data;

[0068] Step 13: Entity classification and entity relationship construction are performed on the architecture layer, and knowledge fusion of different entities is performed using hypernode links and hyperedge links to obtain the initial scene knowledge graph. The scene knowledge graph includes a semantic layer, a data layer, and a scene layer.

[0069] Step 14: Use a hierarchical storage architecture to store each semantic entity in the semantic layer as a scene knowledge graph node, store the original data in the data layer, and store the data in the scene layer in the form of an attribute table. Use a unique identifier and index mechanism to associate the data in the semantic layer, data layer, and scene layer. When the spatiotemporal data of the target city changes, the initial scene knowledge graph is dynamically updated to obtain a scene knowledge graph.

[0070] Specifically, spatiotemporal data, namely the spatiotemporal data of the target city, comes from multiple channels, such as urban statistical yearbooks published by relevant government departments, project documents of urban planning and construction departments, and environmental data reports provided by environmental monitoring agencies; spatiotemporal data covers a longer time period, such as records of city-related data over the past ten years or even longer, which enables the analysis of urban health indicators in time series, such as the growth trend of urban invention patents, changes in economic development speed, etc.; from the spatial dimension, spatiotemporal data covers different areas of the target city, which helps to analyze the differences and distribution patterns of urban health indicators in different regions, such as the density of urban greenways and differences in construction levels in different regions.

[0071] Specifically, business demand data include knowledge-based assessment data, spatiotemporal analysis data, and diversified evidence data. Among them, knowledge-based assessment data include descriptive, diagnostic, predictive, and programmatic knowledge for greenway construction. This knowledge comes from urban planning documents, expert opinions, and analysis of historical data, and aims to provide theoretical support and decision-making basis for greenway construction. Data acquisition methods include literature research, expert interviews, and historical data mining, which are highly accurate and can provide strong support for the scientific planning and management of greenway construction. Spatiotemporal analysis data include urban greenway density, per capita density length, average 1-hour walking greenway accessibility, average 5-minute driving greenway accessibility, etc. These data are mainly obtained through field measurements, geographic information system analysis, and remote sensing technology, and have relatively good accuracy. High spatiotemporal resolution and accuracy can effectively reflect the current status and dynamic changes of greenway construction; diversified evidence data include geographic information data, population data, traffic data, urban planning documents and multimedia data. Geographic information data include topographic maps, road networks, rivers and lakes, greenway locations, greenway lengths, surrounding environments, etc. These data are mainly obtained through satellite remote sensing and field measurements; population data include demographic data such as population distribution and population density in the areas where greenway construction is located; traffic data include the time and location of accidents, etc. Urban planning documents include urban master plans, special plans for greenway construction, etc., which are used to provide document data on the background, goals and specific implementation plans for greenway construction; multimedia data include satisfaction surveys of citizen feedback, online video reports, etc.

[0072] Specifically, the architecture layer in step 12 is built in a top-down manner, with the following process:

[0073] First, based on the business demand data for physical examinations in target cities, we conduct an in-depth analysis of business processes, scenario elements, and associated semantics.

[0074] Then, based on the target city's spatiotemporal data, a multi-level expression is used to describe the semantics of scenario modeling, including the semantic layer, scenario layer, and data layer. The semantic layer involves the knowledge content of business requirements such as the theme, connotation, knowledge points, and indicators of urban health checks. The data layer describes the spatiotemporal data generated during the monitoring, diagnosis, early warning, and maintenance phases of urban health checks. The scenario layer describes the data and knowledge sets required for urban health check application scenarios.

[0075] Finally, we use hyper-node links and hyper-edge links to connect complex data and knowledge in application scenarios to build a hypergraph-based architecture layer.

[0076] Most preferably, the architectural layers are formalized as:

[0077] ;

[0078] ;

[0079] ;

[0080] ;

[0081] in, Represents the architectural layer, Represents a semantic set, Represents the topic node, Represents the connotation node, represents a knowledge node, Represents an indicator node, Represents a data set, Indicates the working stage of the city physical examination. Indicates the data type, Indicates the source of data. Represents data attributes, Represents the relationship between data, Represents a collection of application scenarios, Indicates the type of application scenario. Represents the hierarchical structure of data in the urban physical examination process, , , Both represent the time and space scales, Indicates the physical examination stage, Indicates the physical examination differentiation characteristics, represents a data subgraph consisting of a set of nodes and edges, , Represents a set of feature identifiers of a subgraph in different dimensions (such as subject dimension, knowledge dimension, indicator dimension, etc.). Indicates the attribute name of the subgraph, Representing high-order relations in the data hypergraph, , Represents spatial relationships, Indicates time relationship. Indicates the relationship between the physical examination stages, Represents the physical examination feature relationship, Represents dimensional relationships.

[0082] Specifically, the initial scene knowledge graph in step 13 is also created using a bottom-up approach. The specific process is as follows:

[0083] First, to adapt to the application scenario of urban physical examinations, the embodiment of the present invention divides entities into three categories: knowledge system, data system, and scenario model. Among them, the knowledge system constructs a semantic transmission link with "theme-content-knowledge point-indicator" as the backbone. The data system contains spatiotemporal data generated in the monitoring, diagnosis, early warning and maintenance stages. The scenario model includes scenario ID, time and space entities.

[0084] Then, named entity recognition and attribute extraction are performed on these entities to establish relationships between them, such as semantic relationships such as causality, complementarity, correlation, and hierarchy; data relationships such as expression, operation, analysis, and support; and spatiotemporal relationships such as connection, sequence, topology, orientation, and distance.

[0085] Finally, according to the application scenario requirements, knowledge fusion is performed using the hypernode link and hyperedge link hierarchy, data subgraphs, and high-order relationships to merge the same or similar information, delete erroneous or redundant information, and obtain the initial scenario knowledge graph.

[0086] In order to ensure the timeliness of the scenario knowledge graph, the embodiment of the present invention establishes a city physical examination knowledge graph storage and dynamic update mechanism. In terms of storage, a layered storage architecture is adopted, combined with the neo4j graph data integration storage capabilities and various database management storage capabilities.

[0087] Specifically, this storage architecture stores semantic layer data, data layer raw data, and scenario layer application data separately, establishing close associations through unique identifiers and indexing mechanisms. The specific process is as follows:

[0088] First, various semantic entities in the semantic layer are stored as nodes in the knowledge graph. Each node has a unique identifier in neo4j and stores the attributes and relationship information related to the entity.

[0089] Secondly, the raw data in the data layer is stored in the appropriate file system or key-value storage according to the data type and source. At the same time, data nodes and relationships are created in the knowledge graph. By establishing a reference link to the original data storage location in neo4j, the raw data can be accurately obtained and processed when needed.

[0090] Then, for the data of the scene layer, combined with application requirements, scene-related configuration information (such as scene type, time and space range in scene pre-definition, etc.) and user interaction data (such as user operation records in scene visualization) are stored in the file system or relational database in the form of attribute tables. At the same time, scene nodes and relationships are created in neo4j, and the data and knowledge in the attribute table are associated with the scene nodes.

[0091] In the embodiment of the present invention, in terms of updating the initial scenario knowledge graph, a strategy combining incremental update, full update, and version update is adopted. The specific process is as follows:

[0092] When the spatiotemporal data of the target city changes, incremental updates are used to adjust the entities, relationships, and attributes in the initial scenario knowledge graph based on its existing structure and semantic rules. For newly added or modified data, the existing structure and semantic rules of the initial scenario knowledge graph are used to quickly locate the relevant entities, relationships, and attributes and make local adjustments, such as adding new classes, attributes, or relationships.

[0093] After performing a conflict or consistency check on the adjusted initial scenario knowledge graph, the scenario knowledge graph is obtained. The consistency check includes data type consistency (ensuring that the format of the new data is compatible with the existing data, such as a unified time format and consistent numerical units), semantic consistency (newly added relationships and attributes conform to the established semantic logic of the knowledge graph, such as the reasonable association between greenway construction and related accessibility indicators), and logical consistency (avoiding contradictory information, such as inconsistent infrastructure types in different data sources for the same area). If a conflict is found, semantic rules will be used to resolve the conflict to ensure the consistency and coherence of the updated knowledge graph.

[0094] When the target city's spatiotemporal data needs to be fully updated, the architecture layer and the initial scenario knowledge graph are updated and stored based on the updated target city's spatiotemporal data, including: rebuilding the architecture layer and the initial scenario knowledge graph based on the updated data source and business needs, re-sorting the themes, connotations, knowledge points, indicators and data of the city physical examination, re-identifying and extracting entities, relationships and attributes, and organizing and storing them according to new standards, updating and recording the status changes of the knowledge graph before and after each full update, and obtaining the scenario knowledge graph by establishing version management, tracing back and analyzing the development trajectory of the knowledge graph.

[0095] Specifically, if Figure 4 As shown, step 2 includes:

[0096] Step 21: parse the target city's spatiotemporal data and scene knowledge graph to obtain multiple material elements, and integrate and manage all the material elements to form a scene material library;

[0097] Specifically, in order to organize the large amount of data and knowledge such as indicator analysis, evaluation and evidence collection generated in the urban physical examination business into a unified scene model, the embodiment of the present invention uses material graphics as the main modeling unit, reads the target city's spatiotemporal data, such as urban planning documents, geographic information databases, environmental monitoring reports, socio-economic statistical yearbooks, etc., extracts texts, images, and maps related to the urban scene, and converts them into corresponding material graphics. By parsing the knowledge graph entities, the attributes and relationship information related to the urban physical examination scene are extracted and converted into material graphics to obtain multiple material graphics. Through this method, information such as spatiotemporal statistics, spatial analysis, indicator analysis, "descriptive-diagnostic-predictive-scheme" knowledge, as well as graphic symbols, cross-media, geographic information, policy report evidence and other material graphics are encapsulated, allowing the assembly of "spatiotemporal analysis-knowledge-based evaluation-diversified evidence" scene materials. By integrating and managing these material graphics, a scene material library is formed to support scene models and dynamic updates.

[0098] Step 22: Decompose the scene in the scene knowledge graph to obtain multiple basic modules. Combine semantic constraints to perform element mapping on all basic modules, all material elements in the scene material library, and semantic relationships in the scene knowledge graph to obtain multiple mapping results. All mapping results are assembled to obtain a scene model.

[0099] Step 23, dynamically evolve the scenario model through scenario narrative logic modeling to obtain the evolution result, transform the evolution result through expression processing to obtain the initial city physical examination scenario, and update the initial city physical examination scenario through the dynamic update mechanism to obtain the city physical examination scenario.

[0100] Specifically, step 22 includes:

[0101] First, based on the knowledge graph of urban physical examination scenarios, the scenario is semantically analyzed and decomposed into multiple basic modules, which are formally expressed as follows:

[0102] ;

[0103] in, Represents the basic module, represents the decomposition function, Represents the theme node, which identifies the main themes in the urban health check scene, such as ecological livability, convenient transportation, innovation and vitality, etc. Represents the connotation node, and extracts the connotation of each theme, that is, the connotation covered by the theme. For example, the connotation of the ecological livable theme includes green and low-carbon development, ecological environment protection, living space, etc. Represents knowledge nodes, and the knowledge points related to the connotation are sorted out. For example, the knowledge points under the connotation of green and low-carbon development include greenway construction, energy conservation and emission reduction, water resource utilization, etc. Represents the indicator node, which determines the indicator used to evaluate each knowledge point, such as greenway density, greenway walking accessibility, and greenway vehicle accessibility. Represents a data set, which is obtained by classifying scene-related data, including values, tables, text, events, etc.

[0104] Next, we perform primitive mapping based on semantic constraints, which can be formally expressed as follows:

[0105] ;

[0106] in, Represents the mapped primitives, represents the primitive mapping function, Represents material primitives, used to fill basic modules;

[0107] Using the above formula, we extract scene-related primitives from the scene material library and map the extracted primitives to scene templates used to organize and display the primitives. Based on the semantic relationships, we lay out the scene templates and adjust the positional relationships, colors, and sizes between the primitives.

[0108] Finally, assemble the scene model in the scene template , which is formally expressed as:

[0109] ;

[0110] in, Indicates the initial city physical examination scenario, Represents visualization operations, Represents the timeline, Represents a geospatial scene.

[0111] Since the urban health check scenario narrative involves the dynamic evolution of geographic space scenarios and follows a specific logic, it is necessary not only to identify events, describe the detailed processes corresponding to the events, express the states contained in the processes and the people / events in the states, but also to use a spatiotemporal framework to organize and present information. Therefore, the scenario model is dynamically evolved through scenario narrative logic modeling to obtain the evolution results, including:

[0112] First, identify key events in a certain road section with potential dangers during the city physical examination, such as traffic accidents and greenway construction;

[0113] For each identified event, describe its occurrence, development, and end. For example, a traffic accident event may begin at a certain time, with traffic congestion gradually increasing, posing a danger to both people and vehicles. Finally, at a certain point in time, a dedicated bicycle lane may be opened on the hazardous road section to alleviate these problems.

[0114] In the process description, express each state and the people and things it contains. For example, in the process of vehicle congestion, the state involves huge traffic flow and limited green travel. The limited green travel state involves a large local population and a surge in motor vehicles.

[0115] The spatiotemporal framework is used to organize and present scene information and obtain evolution results. The spatiotemporal framework includes geographic space scenes and time series. The geographic space scenes can be two-dimensional or three-dimensional city maps or real-life images, showing the specific location of the event. The time series records the state changes of the event at different time points.

[0116] Specifically, the evolution results are transformed through expression processing to obtain the initial urban physical examination scenario, including:

[0117] The scene narrative logic model is transformed into an understandable and operable scene information form through expression processing. The expression processing is formally expressed as:

[0118] ;

[0119] ;

[0120] ;

[0121] in, Represents a sorted organization, Represents the spatial layout, Represents the semantic conversion method, Indicates the semantics, Represents the process, Indicates an event, Indicates status, Indicates a place;

[0122] Using the above formula, events, processes, and states are organized into a timeline in chronological order, making it easier for decision makers to understand the development process and status of events. Spatial entities are organized and laid out according to geographic spatial scenarios. The semantics are then converted into understandable text, symbols, or non-physical expressions.

[0123] Specifically, the initial city physical examination scenario is updated through a dynamic update mechanism to obtain a city physical examination scenario, including:

[0124] By formula Dynamically update the material primitives to obtain the updated scene material library, where: Represents the material element, Indicates the parsing method. represents the target city’s spatiotemporal data, Represents scene knowledge graph;

[0125] By formula Dynamically update the scene model to obtain an updated scene model, where: Represents the scene model, represents a composition operation, Indicates the mapped primitives;

[0126] By formula The initial city physical examination scene is updated to obtain the city physical examination scene, where: Indicates the initial city physical examination scenario, Represents visualization operations, Represents the timeline, Represents a geospatial scene.

[0127] Specifically, step 3 includes:

[0128] Step 31, enhancing the semantics of objects in the urban physical examination scene based on the combination of static and dynamic visual variables to obtain an enhanced urban physical examination scene;

[0129] Step 32: Prioritize the data in the enhanced urban physical examination scenario based on the importance and business requirements of the data in the urban physical examination scenario to obtain a priority result;

[0130] Step 33: Schedule and present the data of the enhanced urban physical examination scene according to the division results to obtain a dynamic expression result of the urban physical examination scene.

[0131] Because traditional visual variables, such as shape, size, color, brightness, orientation, and texture, primarily focus on static features and are unable to fully express the dynamic characteristics and behavioral relationships of geographic phenomena, urban health scene visualization requires the use of visual variables to transform complex urban health scene into intuitive and easy-to-understand visualization information. To overcome this limitation, dynamic visual variables such as time, frequency, duration, synchronization, and sequence are proposed by introducing the temporal dimension. These variables can more accurately and intuitively reflect the state and characteristics of spatial phenomena. Therefore, the combination of "dynamic" and "static" visual variables enhances understanding and effectively highlights target information.

[0132] Most preferably, the formal expression for enhancing the semantics of objects in the urban physical examination scene based on the combination of static and dynamic visual variables is:

[0133] ;

[0134] in, represents a set of visual variables, represents dynamic visual variables, represents static visual variables, represents the scene features for enhanced representation, Represents the feature information of scene objects, Represents spatial location information, Indicates attribute information. Represents related information, such as causality and time.

[0135] Specifically, the static part is a scene assessment story composed of multiple knowledge points and relationships. It enhances the semantic expression of scene objects such as text descriptions and statistical charts through static visual variables (such as shape, size, color, etc.). The dynamic part is based on the scene assessment storyline and introduces dynamic visual variables (such as time, frequency, duration, etc.) to show the dynamic changes of scene objects such as geographic space or video media; by combining the two (such as direct rendering, dynamic rendering), scene dynamic symbols (such as protrusion, movement, flashing) are formed to make the urban physical examination scene information presentation more comprehensive and vivid.

[0136] Since the urban health examination scenario involves not only static data such as multi-scale geographic scenes, text descriptions, and statistical charts, but also dynamic data formed by the combination of video media, multi-scale geographic scenes, and dynamic symbols, static data does not change frequently over time, while dynamic data has a time dimension and changes frequently, requiring real-time updates and displays. Based on the importance and business needs of the data in the urban health examination scenario, the formal expression of data priority in the enhanced urban health examination scenario is as follows:

[0137] ;

[0138] ;

[0139] in, Indicates high priority data, Represents the feature information of geographic scene objects, Represents the feature information of the video object, Indicates the characteristic information of dynamic symbols, Indicates low priority data, Represents the characteristic information of the icon object. Represents the feature information of a text object.

[0140] For high-priority data, such as key geographic scenes and real-time surveillance videos, preloading is used to ensure that users can quickly access information when viewing. For low-priority data, such as detailed statistical charts and text descriptions, on-demand loading is used to ensure that business personnel obtain concise and useful information.

[0141] Most preferably, the formal expression for scheduling and presenting the data of the enhanced urban physical examination scene according to the division results is:

[0142]

[0143] in, object feature data representing the scene, Indicates the scene scheduling and presentation method, Indicates the priority division result, Indicates the data display method. Indicates the publishing method.

[0144] Specifically, in terms of display mode, the embodiment of the present invention renders static data as the background layer and dynamic data as the foreground layer to avoid visual confusion caused by information superposition. Its formal expression is:

[0145] ;

[0146] ;

[0147] in, Represents background layer data, providing basic geographic, text, and statistical information. Represents foreground layer data and displays real-time change information. Indicates the rendering method.

[0148] Select the appropriate transmission protocol and publishing method based on the data volume, network environment, and business requirements. The formal expression is:

[0149] ;

[0150] in, Indicates publishing tile map data. Indicates publishing dynamic map data. Indicates publishing dynamic real-time video data. Indicates publishing static text and chart data.

[0151] Use the above formula to publish the scene object, specifically:

[0152] For multi-scale geographic scene data, use the Web Map Tile Service protocol WMTS and visualization libraries (such as Leaflet and OpenLayers) to publish it as a tile map service, and load and render a map with a fixed window range through a URL;

[0153] Dynamic map services are published using the Web Map Service protocol WMS and ArcGIS Server. Interactive maps are loaded and rendered through the RESTful API to display urban geographic spatial elements with different symbols and layers. For dynamic data with high real-time requirements (such as real-time video data), the HLS protocol is used for real-time push. For static data such as small-scale text and statistical data, the HTTP protocol and visualization libraries (such as Echarts and D3.js) are used to convert the data into graphical and chart visualization forms.

[0154] In this embodiment of the present invention, a southern city with a complex urban structure and diverse urban functional areas is used as the experimental area. Urban greenway construction is used as an application scenario for urban health check. Three typical urban health check scenarios of urban greenway construction are selected to verify the effectiveness of the provided method. The specific process is as follows:

[0155] Collect and process the spatiotemporal data required for a greenway construction scenario in a southern city, and use this spatiotemporal data to obtain business demand data, such as knowledge-based assessment data, spatiotemporal analysis data, and diversified evidence data.

[0156] Urban greenway health check scenarios encompass multiple scales, such as the overall city status at the macro scale, regional conditions at the meso scale, and specific events at the micro scale. Each scenario integrates the required business knowledge and the occurrence, development, and conclusion of events to form a complete assessment story.

[0157] Extract scene elements of different scale structures:

[0158] First, the overall greenway construction situation of the city is extracted at the macro scale, including themes, connotations, total length of greenways, coverage areas, and per capita greenway density indicators.

[0159] Secondly, at the meso-scale, we extract the greenway construction situation in specific areas, such as the distribution of greenways in Bio-Island, greenway repair and renovation in Huangpu District;

[0160] Secondly, detailed information about specific events is extracted at the micro scale, such as the construction process, usage, and resident feedback of a greenway. By extracting material elements (such as text, images, maps, etc.), the material elements are mapped to the scene template;

[0161] Finally, layout and adjustment are carried out according to semantic relationships to form scene models with different scale structures. Among them, the macro scale represents an overview of the overall greenway construction in the city, the meso scale represents a detailed map of the greenway construction in a specific area, and the micro scale represents the specific greenway construction and use process.

[0162] In order to provide a background for greenway construction and the improvement effect of the reconstruction of "risky" road sections, the scene dynamic expression is carried out from three scales: macro, meso and micro:

[0163] First, GIS and vector analysis maps are used to display the Guangzhou greenway network pattern at the macro scale, and three-dimensional maps and visualization charts are used to display descriptive knowledge such as the construction status of park greenways at the meso scale.

[0164] Then, problem diagnosis was conducted at a microscale. By identifying key events in the greenway construction on the Gangding Road section of the area, such as traffic accidents and greenway openings, the accident locations and greenway opening locations were marked on the map, and key events were highlighted through dynamic symbols (such as flashing text labels and changing icon colors).

[0165] Finally, through interactive scene switching, the solution-based knowledge of "improving the safety hazards of mixed traffic of people and vehicles through the opening of greenways" is dynamically updated.

[0166] In order to provide construction background, problem diagnosis and optimization measures for existing greenways, the scene dynamics are expressed at both the meso- and micro-scales:

[0167] First, at the mesoscale, three-dimensional maps and vector analysis diagrams are used to display descriptive knowledge such as the overall pattern of the existing greenway network in the bio-island.

[0168] Secondly, problem diagnosis is conducted at the micro-scale. Key issues that disrupt the greenway around the island, such as building obstructions and uneven road widths, are identified. The locations of greenway interruptions are marked on the map, and dynamic symbols (e.g., greenway color changes from light green to dark green) are used to represent the interruptions.

[0169] Finally, by interactively switching the existing greenway scenes, the “descriptive-diagnostic” knowledge update is achieved.

[0170] To provide context for the construction of new greenways on bridges and the effectiveness of ecological transportation optimization, the following scene dynamics are presented at both the meso- and micro-scales:

[0171] First, at the mesoscale, three-dimensional maps and visualization charts are used to display the cross-shore transportation pattern and analyze the current construction status and existing problems;

[0172] Then, at the micro-scale, identify pedestrian bridge construction cases, mark the locations of the newly added greenways on the map, and use dynamic symbols (such as green pedestrian lines) to express the new parts, visually showing the location and connection relationship of the new greenways in the regional transportation network;

[0173] Finally, dynamic scene updates are achieved through interactive scene switching.

[0174] The embodiment of the present invention uses a questionnaire survey to quantitatively analyze the cognitive evaluation of business personnel on urban greenway scenes. 80 participants in the field of urban planning participated in this comparative experiment and were randomly divided into two groups: Group A and Group B. The 46 participants in Group A were asked to read a text report, and the 34 participants in Group B were asked to watch a dynamic scene expression page. In order to evaluate the reliability and validity of the questionnaire, Cronbach's Coefficient (Cronbach's ) and Kaiser-Meyer-Olkin (KMO) test. It is generally believed that Cronbach The coefficient reflects the consistency and reliability of the questionnaire results, with a value greater than 0.8 indicating high reliability. The KMO test assesses the construct validity of the questionnaire, with a value greater than 0.7 indicating acceptable validity. Furthermore, this embodiment of the present invention uses metrics such as mean, quartiles, and standard deviation to analyze feedback. The mean reflects the user's level of understanding; the quartiles are used to analyze the central tendency of user feedback; and the standard deviation indicates the dispersion of the test results and the stability of the pre-set questions.

[0175] After receiving 80 questionnaires, the reliability and validity analysis of the questionnaire data was carried out. The value was determined to be 0.85, which is greater than 0.8; the KMO value was 0.76, which is close to 0.8. These statistics show that the data and results collected through the questionnaire are true, valid and reliable.

[0176] The present invention uses indicators such as mean, quartiles, and standard deviation for analysis, and uses the Mann–Whitney U test to compare the differences between Group A and Group B. u and z represent statistics, and the P value represents the corresponding probability, resulting in the following Table 1:

[0177] Table 1 Statistical test of experimental results

[0178] ;

[0179] As can be seen from Table 1 above, when the P value is less than 0.01, it indicates a high degree of statistical significance. In the three evaluation indicators of practicality, understandability and aesthetics, the P values are all 0.000, all less than 0.01, which shows that there are significant differences between Group A and Group B in these three indicators.

[0180] The embodiment of the present invention obtains the business demand data of the target city based on the target city's spatiotemporal data, and constructs a scenario knowledge graph based on the target city's spatiotemporal data and the business demand data; the target city's spatiotemporal data and the scenario knowledge graph are parsed to obtain parsing results, and the parsing results are reorganized by combining semantic constraints and graphic element combinations to drive scene generation and update with the help of scenario narrative logic to obtain a city physical examination scene; the semantics of objects in the city physical examination scene are enhanced in a dynamic and static manner to obtain an enhanced city physical examination scene, and the enhanced city physical examination scene is presented with data priority scheduling to obtain a dynamic expression result of the city physical examination scene; compared with the prior art, the embodiment of the present invention For example, by introducing knowledge graphs to construct scene knowledge graphs to deeply associate urban spatiotemporal data and business demand data, the problem of disconnection between business and data in existing technologies is solved. Semantic constraints and graphic element combinations are combined to guide the reorganization of target city spatiotemporal data and scene knowledge graph analysis results. Scene generation and updating are driven by scene narrative logic, which solves the problem of insufficient adaptability of dynamic scene expression. The semantics of objects in the urban physical examination scene are enhanced in a dynamic and static combination method to obtain an enhanced urban physical examination scene, and data is prioritized for the enhanced urban physical examination scene, which significantly improves the visualization results of the urban physical examination scene and solves the problem of relatively single visualization presentation.

[0181] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A knowledge-guided dynamic expression method for urban physical examination scenarios, characterized by: include: Step 1: Acquire the target city's business demand data based on the target city's spatiotemporal data, and construct a scenario knowledge graph based on the target city's spatiotemporal data and the business demand data. The business demand data includes knowledge-based assessment data, spatiotemporal analysis data, and diversified evidence data. Step 2: parse the target city's spatiotemporal data and the scenario knowledge graph to obtain parsing results, combine semantic constraints with graph element combinations to guide the reorganization of the parsing results, and use scenario narrative logic to drive scenario generation and update to obtain a city physical examination scenario; Step 3: Enhance the semantics of objects in the urban physical examination scene by combining static and dynamic methods to obtain an enhanced urban physical examination scene, and perform data-priority scheduling and presentation on the enhanced urban physical examination scene to obtain a dynamic expression result of the urban physical examination scene.

2. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 1 is characterized in that: The step 1 comprises: Step 11: Acquire the business demand data of the target city based on the spatiotemporal data of the target city; Step 12: constructing a hypergraph-based architecture layer according to the business demand data and the target city spatiotemporal data; Step 13: Entity classification and entity relationship construction are performed on the architecture layer, and knowledge fusion of different entities is performed using hypernode links and hyperedge links to obtain an initial scene knowledge graph, which includes a semantic layer, a data layer, and a scene layer. Step 14: Use a hierarchical storage architecture to store each semantic entity in the semantic layer as a scene knowledge graph node, store the original data in the data layer, and store the data in the scene layer in the form of an attribute table. Associate the data in the semantic layer, the data layer, and the scene layer through a unique identifier and index mechanism. When the spatiotemporal data of the target city changes, dynamically update the initial scene knowledge graph to obtain a scene knowledge graph.

3. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 2 is characterized in that: The formal expression of the architecture layer is: in, Represents the architectural layer, Represents a semantic set, Represents the topic node, Represents the connotation node, represents a knowledge node, Represents an indicator node, Represents a data set, Indicates the working stage of the city physical examination. Indicates the data type, Indicates the source of data. Represents data attributes, Represents the relationship between data, Represents a collection of application scenarios, Indicates the type of application scenario. Represents the hierarchical structure of data in the urban physical examination process, represents a data subgraph consisting of a set of nodes and edges, Represents higher-order relations in a data hypergraph.

4. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 2 is characterized in that: When the target city's spatiotemporal data changes, the initial scene knowledge graph is dynamically updated to obtain a scene knowledge graph, including: When the spatiotemporal data of the target city changes, based on the existing structure and semantic rules of the initial scene knowledge graph, the entities, relationships and attributes in the initial scene knowledge graph are adjusted by incremental update, and the adjusted initial scene knowledge graph is checked for conflicts or consistency to obtain a scene knowledge graph; When the target city's spatiotemporal data needs to be comprehensively revised, the architecture layer and the initial scenario knowledge graph are updated and stored based on the revised target city's spatiotemporal data to obtain a scenario knowledge graph.

5. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 4 is characterized in that: The step 2 includes: Step 21: parse the target city's spatiotemporal data and the scene knowledge graph to obtain multiple material graphics elements, and integrate and manage all the material graphics elements to form a scene material library; Step 22: Decompose the scene in the scene knowledge graph to obtain multiple basic modules, perform element mapping on all basic modules, all material elements in the scene material library, and semantic relationships in the scene knowledge graph in combination with semantic constraints to obtain multiple mapping results, and assemble them based on all the mapping results to obtain a scene model; Step 23, dynamically evolve the scenario model through scenario narrative logic modeling to obtain an evolution result, transform the evolution result through expression processing to obtain an initial city physical examination scenario, and update the initial city physical examination scenario through a dynamic update mechanism to obtain a city physical examination scenario.

6. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 5 is characterized in that: The initial city physical examination scenario is updated through a dynamic update mechanism to obtain a city physical examination scenario, including: By formula Dynamically update the material primitives to obtain an updated scene material library, wherein: Represents the material element, Indicates the parsing method. represents the target city’s spatiotemporal data, Represents scene knowledge graph; By formula The scene model is dynamically updated to obtain an updated scene model, wherein: Represents the scene model, represents a composition operation, Indicates the mapped primitives; By formula The initial city physical examination scene is updated to obtain a city physical examination scene, wherein: Indicates the initial city physical examination scenario, Represents visualization operations, Represents the timeline, Represents a geospatial scene.

7. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 5 is characterized in that: Step 3 includes: Step 31, enhancing the semantics of objects in the urban physical examination scene based on the combined visual variables of movement and stillness to obtain an enhanced urban physical examination scene; Step 32: Prioritize the data in the enhanced urban physical examination scenario according to the importance and business requirements of the data in the urban physical examination scenario to obtain a priority result; Step 33: Schedule and present the data of the enhanced urban physical examination scene according to the division result to obtain a dynamic expression result of the urban physical examination scene.

8. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 7 is characterized in that: The formal expression of enhancing the semantics of objects in the urban physical examination scene based on the combination of dynamic and static visual variables is: in, represents a set of visual variables, represents dynamic visual variables, represents static visual variables, represents the scene features for enhanced representation, Represents the feature information of scene objects, Represents spatial location information, Indicates attribute information. Indicates related information.

9. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 8 is characterized in that: According to the importance and business requirements of the data in the urban physical examination scenario, the formal expression of the priority classification of the data in the enhanced urban physical examination scenario is as follows: in, Indicates high priority data, Represents the feature information of geographic scene objects, Represents the feature information of the video object, Indicates the characteristic information of dynamic symbols, Indicates low priority data, Represents the characteristic information of the icon object. Represents the feature information of a text object.

10. The knowledge-guided dynamic expression method for urban physical examination scenarios according to claim 9 is characterized in that: The formal expression for scheduling and presenting the data of the enhanced urban physical examination scene according to the division result is: in, object feature data representing the scene, Indicates the scene scheduling and presentation method, Indicates the priority division result, Indicates the data display method. Indicates the publishing method.

Images