GIS layer analysis method and system based on automatic image combination and analysis

Abstract

The invention discloses a GIS layer analysis method and system based on automatic graph combination and analysis, and relates to the related field of geographic information systems, and the method comprises the steps: accessing a multi-source dynamic layer data stream updated at a frequency greater than a preset frequency in real time; on the basis of the multi-source dynamic layer data flow, space elements serve as nodes, association rules serve as directed edges, and a space-time causal knowledge graph is constructed; performing multi-layer joint simulation according to the space-time causal knowledge graph, and generating a prediction thematic layer containing a time sequence confidence interval; the dynamic time warping distance between the prediction thematic layer and the real-time monitoring layer is calculated, and when the dynamic time warping distance is larger than a self-adaptive distance threshold value, a difference warning signal bound by space coordinates is generated. The technical problem that existing GIS layer analysis based on automatic graph combination and analysis is insufficient in prediction precision and reliability is solved, and the technical effect of improving prediction precision and reliability is achieved.

Description

Technical Field

[0001] This application relates to the field of geographic information systems, and particularly to a GIS layer analysis method and system based on automatic map synthesis and parsing. Background Art

[0002] In fields such as urban planning, environmental management, and emergency response, real-time monitoring and prediction of changes in geospatial data can effectively reduce potential risks and improve the efficiency and safety of urban operations. Currently, traditional geographic information system (GIS) technology combined with simple statistical analysis methods is mainly used to process and predict geospatial data. These methods rely on static data and single-time-scale analysis, suffering from problems such as low data update frequency, lack of spatio-temporal correlation analysis, and insufficient early warning capabilities, resulting in insufficient prediction accuracy and reliability.

[0003] In the current related technologies, there are technical problems of insufficient prediction accuracy and reliability in GIS layer analysis based on automatic map synthesis and parsing. Summary of the Invention

[0004] This application provides a GIS layer analysis method and system based on automatic map synthesis and parsing. By first accessing a multi-source dynamic layer data stream with high-frequency updates in real time, then constructing a spatio-temporal causal knowledge graph with spatial elements as nodes and association rules as directed edges, then conducting multi-layer joint simulation based on this graph to generate a prediction thematic layer with a time series confidence interval, and finally calculating the dynamic time warping distance between the prediction thematic layer and the real-time monitoring layer, and generating a differential warning signal bound to spatial coordinates when the dynamic time warping distance is greater than an adaptive threshold, etc., the technical effect of improving the prediction accuracy and reliability is achieved.

[0005] This application provides a GIS layer analysis method based on automatic map synthesis and parsing, including: accessing a multi-source dynamic layer data stream updated at a frequency greater than a preset frequency in real time; based on the multi-source dynamic layer data stream, constructing a spatio-temporal causal knowledge graph with spatial elements as nodes and association rules as directed edges; conducting multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval; calculating the dynamic time warping distance between the prediction thematic layer and the real-time monitoring layer, and generating a differential warning signal bound to spatial coordinates when the dynamic time warping distance is greater than an adaptive distance threshold.

[0006] In a possible implementation, based on the multi-source dynamic layer data stream, using spatial features as nodes and association rules as directed edges, a spatio-temporal causal knowledge graph is constructed, and the following processing is performed: Extract the static spatial features and corresponding attribute information in the multi-source dynamic layer data stream to construct the static nodes of the knowledge graph; Extract the dynamic spatial features and corresponding spatio-temporal change characteristics in the multi-source dynamic layer data stream to construct the dynamic nodes of the knowledge graph; Construct association rules according to the interaction relationship between the static spatial features and the dynamic spatial features, and establish the directed edges of the knowledge graph based on the association rules; Construct a spatio-temporal causal knowledge graph based on the static nodes, the dynamic nodes, and the directed edges.

[0007] In a possible implementation, according to the interaction relationship between the static spatial features and the dynamic spatial features, association rules are constructed, and the directed edges of the knowledge graph are established based on the association rules. The following processing is performed: If the spatial co-occurrence probability of any two static spatial features is greater than a first preset threshold, a first dynamic association rule is generated, and a first directed edge of the knowledge graph is established based on the first dynamic association rule; If the time lag correlation of any two dynamic spatial features is greater than a second preset threshold, a second dynamic association rule is generated, and a second directed edge of the knowledge graph is established based on the second dynamic association rule; If the dependence correlation between any one static spatial feature and any one dynamic spatial feature is greater than a third preset threshold, a third dynamic association rule is generated, and a third directed edge of the knowledge graph is established based on the third dynamic association rule.

[0008] In a possible implementation, multi-layer joint simulation is performed according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval. The following processing is performed: Perform time discretization processing on the spatio-temporal causal knowledge graph to generate a time propagation sequence; Based on the time propagation sequence, starting from the states of each node at the current moment, perform state transfer simulation along the directed edges to obtain a first prediction thematic layer; Based on the first prediction thematic layer, perform simulation of adding random noise to the dynamic nodes to obtain a second prediction thematic layer; Add a confidence interval to the second prediction thematic layer to obtain a prediction thematic layer including a time series confidence interval.

[0009] In a possible implementation, a confidence interval is added to the second predicted thematic layer to obtain a predicted thematic layer containing a time series confidence interval, and the following processing is performed: Obtain the basic number of simulation runs N, where N ≥ 1000; perform multi-layer joint simulation according to the basic number of simulation runs N to obtain N second predicted thematic layers; for the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers, calculate the probability distribution function using a Gaussian kernel density; for non-critical regions, take the 5% - 95% quantiles of the probability distribution function as the 90% confidence interval, and for critical regions, take the 10% - 90% quantiles of the probability distribution function as the 80% confidence interval; calculate the mean value of the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers to obtain a second mean predicted thematic layer; add the confidence interval to the second mean predicted thematic layer to obtain a predicted thematic layer containing a time series confidence interval.

[0010] In a possible implementation, calculate the dynamic time warping distance between the predicted thematic layer and the real-time monitoring layer, and perform the following processing: perform preprocessing for time series alignment on the predicted thematic layer and the real-time monitoring layer to obtain an aligned layer pair; perform multi-scale dynamic warping on the aligned layer pair to obtain a warped layer pair; calculate the cosine similarity and root mean square error of the warped layer pair; calculate the weighted cosine similarity and root mean square error to obtain the dynamic time warping distance.

[0011] In a possible implementation, perform multi-scale dynamic warping on the aligned layer pair to obtain a warped layer pair, and perform the following processing: perform coarse-grained alignment on the aligned layer pair to obtain a coarse-grained aligned layer pair; perform fine-grained optimization on the coarse-grained aligned layer pair to obtain a warped layer pair.

[0012] In a possible implementation, when the dynamic time warping distance is greater than the adaptive distance threshold, generate a difference warning signal bound to spatial coordinates, and perform the following processing: obtain the historical dynamic time warping distance within a preset time range, and take the 90% quantile of the historical dynamic time warping distance as the reference distance threshold; calculate the quality score of the current data, and adaptively adjust the reference distance threshold according to the quality score to obtain the adaptive distance threshold; when the dynamic time warping distance is greater than the adaptive distance threshold, generate a multi-level difference warning signal bound to spatial coordinates according to the distance difference and mark the difference area on the map.

[0013] In a possible implementation, perform the following processing: The multi-source dynamic layer data stream includes: infrastructure sensor data, human activity density heat map, and environmental parameter time series change layer.

[0014] The present application also provides a GIS layer analysis system based on automatic map synthesis and parsing, including: a multi-source dynamic layer data stream access module for real-time accessing multi-source dynamic layer data streams updated at a frequency greater than a preset frequency; a spatio-temporal causal knowledge graph construction module for constructing a spatio-temporal causal knowledge graph based on the multi-source dynamic layer data streams, using spatial elements as nodes and association rules as directed edges; a multi-layer joint simulation module for performing multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval; and a difference warning signal generation module for calculating the dynamic time warping distance between the prediction thematic layer and the real-time monitoring layer, and generating a difference warning signal bound to spatial coordinates when the dynamic time warping distance is greater than an adaptive distance threshold.

[0015] It is intended to first real-time access multi-source dynamic layer data streams updated at a frequency greater than a preset frequency through the GIS layer analysis method and system based on automatic map synthesis and parsing proposed in the present application, then construct a spatio-temporal causal knowledge graph based on the multi-source dynamic layer data streams, using spatial elements as nodes and association rules as directed edges, then perform multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval, and finally calculate the dynamic time warping distance between the prediction thematic layer and the real-time monitoring layer, and generate a difference warning signal bound to spatial coordinates when the dynamic time warping distance is greater than an adaptive distance threshold. The technical effect of improving the accuracy and reliability of prediction is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in the present application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to the need, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0017] Figure 1 It is a schematic flowchart of the GIS layer analysis method based on automatic map synthesis and parsing provided by the embodiments of the present application.

[0018] Figure 2 It is a schematic structural diagram of the GIS layer analysis system based on automatic map synthesis and parsing provided by the embodiments of the present application.

[0019] Description of the reference numerals: multi-source dynamic layer data stream access module 10, spatio-temporal causal knowledge graph construction module 20, multi-layer joint simulation module 30, difference warning signal generation module 40. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The above description is only an overview of the technical solution of this application. In order to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives the specific implementation manners of this application.

[0021] In order to make the purpose, technical solution and advantages of this application clearer, the following will further describe this application in detail with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

[0022] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first / second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0023] The embodiments of this application provide a GIS layer analysis method based on automatic map composition and parsing, as Figure 1 shown, the method includes:

[0024] Step S100, accessing in real time a multi-source dynamic layer data stream updated at a frequency greater than a preset frequency.

[0025] Specifically, a message queue (such as Kafka, RabbitMQ) is used to receive in real time the dynamic layer data streams from different sensors, satellites, Internet of Things devices, etc. The message queue can efficiently process high-frequency data access and support the convergence of multi-source data. Among them, the preset frequency refers to the minimum data update frequency that the system can handle. For example, if the preset frequency is 1 Hz (i.e., updated once per second), the system needs to be able to access in real time data updated once per second or more frequently.

[0026] Develop adapters for different data sources to uniformly convert data in various formats into standard formats (such as GeoJSON, GML) for subsequent processing. Use a stream processing engine (such as Apache Flink, Apache Storm) to clean, format, and perform preliminary screening on real-time data to ensure data integrity and consistency. The stream processing engine can process high-frequency data in real time and support complex event processing logic. Filter out invalid or incorrect data through validation rules (such as data range checks, format validation). Temporarily store the processed data in an in-memory database (such as Redis) for subsequent quick access and processing. The in-memory database can provide low-latency data access and is suitable for real-time analysis scenarios. For large-scale data, use a distributed file system (such as HDFS) for persistent storage to ensure data reliability and scalability.

[0027] For example, assume that it is necessary to access real-time meteorological data from meteorological satellites and environmental monitoring data from ground sensors. By configuring a Kafka cluster, the data from the meteorological satellite is pushed to a Kafka topic at a frequency of once per second, and at the same time, the data from the ground sensors once per second is also pushed to the same topic. The stream processing engine Apache Flink subscribes to this topic, formats the data, removes invalid data, and temporarily stores the processed data in Redis.

[0028] In a possible implementation, step S100 further includes step S110, and the multi-source dynamic layer data stream includes: infrastructure sensor data, human activity density heat map, and environmental parameter time series change layer.

[0029] Specifically, infrastructure sensor data refers to real-time monitoring data from urban infrastructure (such as traffic cameras, bridge sensors, water supply network sensors, etc.). The sensor data is transmitted to a message queue (such as Kafka) in real time through an Internet of Things protocol (such as MQTT, CoAP). Use a stream processing engine (such as Apache Flink) to clean and format the data and extract key information (such as sensor location, monitoring value).

[0030] The human activity density heat map refers to human activity data from mobile devices (such as mobile phone location data, social media check-in data). Obtain data from data providers (such as mobile operators, social media platforms) through an API interface (such as RESTful API) and push it to a message queue. Use geospatial processing tools (such as GDAL, GEOS) to convert the data into a geographic coordinate format (such as GeoJSON) and generate a heat map.

[0031] The environmental parameter time-series change layer refers to real-time environmental data (such as temperature, humidity, air quality index) from weather stations and environmental monitoring stations. Data is obtained from the monitoring stations through protocols such as FTP and HTTP and pushed to the message queue. Time-series processing tools (such as Pandas) are used to align and interpolate the data in terms of time to generate the time-series change layer. By integrating infrastructure sensor data, human activity density heatmaps, and environmental parameter time-series change layers, the fusion of multi-source data is achieved, which can provide a more comprehensive view of the urban operation status and richer data support for urban planning and management.

[0032] Step S200: Based on the multi-source dynamic layer data stream, construct a spatio-temporal causal knowledge graph with spatial elements as nodes and association rules as directed edges.

[0033] Specifically, use GIS software (such as ArcGIS, QGIS) or geospatial processing libraries (such as GDAL, GEOS) to extract spatial elements (such as geographical coordinates, geographical region boundaries, etc.) from the layer data. Extract the spatial information in the data through parsing tools (such as the geopandas library in Python) and convert it into a unified coordinate system (such as WGS84). Use machine learning algorithms (such as association rule mining algorithms Apriori, FP-Growth) to mine the association rules between spatial elements from multi-source data. These algorithms can discover frequent patterns and causal relationships in the data. Combine time-series analysis methods (such as autocorrelation analysis, cross-correlation analysis) to mine the spatio-temporal associations between spatial elements. Use a graph database (such as Neo4j, OrientDB) to construct a spatio-temporal causal knowledge graph, store the spatial elements as nodes and the association rules as directed edges in the graph database. The spatio-temporal causal knowledge graph is used to represent the spatio-temporal causal relationships between spatial elements. Design a dynamic update mechanism for the graph to ensure that new data can be updated to the knowledge graph in a timely manner and maintain the timeliness of the graph.

[0034] For example, extract spatial elements such as temperature and humidity from meteorological data, and extract spatial elements such as air quality index from environmental monitoring data. Use the Apriori algorithm to analyze the association rules between these elements, such as "when the temperature is higher than 30°C and the humidity is lower than 30%, the air quality index will decrease". Store these elements and rules in the Neo4j graph database to form a spatio-temporal causal knowledge graph. When new data is accessed, the nodes and edges in the graph are updated in real time through the stream processing engine.

[0035] In a possible implementation, based on the multi-source dynamic layer data stream, a spatio-temporal causal knowledge graph is constructed with spatial features as nodes and association rules as directed edges. Step S200 further includes step S210 of extracting static spatial features and corresponding attribute information from the multi-source dynamic layer data stream to construct static nodes of the knowledge graph. Specifically, GIS tools (such as ArcGIS, QGIS) or geospatial processing libraries (such as GDAL, GEOS) are used to extract static spatial features from the multi-source dynamic layer data stream, such as buildings, roads, rivers, etc. The attribute information of these features is extracted, such as the height of buildings, the type of roads, the flow rate of rivers, etc. The extracted static spatial features and their attribute information are stored as static nodes of the knowledge graph in a graph database (such as Neo4j). Each static node contains spatial location information (such as longitude and latitude coordinates) and attribute information (such as type, height, length, etc.). For example, building and road information is extracted from the urban planning layer. The geopandas library in Python is used to read the layer data in GeoJSON format, and the coordinates, height, and usage of buildings, as well as the coordinates, type, and length of roads are extracted. This information is stored as static nodes in the Neo4j graph database.

[0036] Step S220, extract dynamic spatial features and corresponding spatio-temporal change characteristics from the multi-source dynamic layer data stream to construct dynamic nodes of the knowledge graph. Specifically, GIS tools or geospatial processing libraries are used to extract dynamic spatial features from the multi-source dynamic layer data stream, such as traffic flow, air quality, population flow, etc. The spatio-temporal change characteristics of these features are extracted, such as time series data, change rates, etc. The extracted dynamic spatial features and their spatio-temporal change characteristics are stored as dynamic nodes of the knowledge graph in the graph database. Each dynamic node contains spatial location information, time series data, and change characteristics. For example, real-time traffic flow data is obtained from traffic cameras, and air quality data is obtained from environmental monitoring stations. The Pandas library is used to process the time series data, and the change rate of traffic flow and the temporal change of air quality are extracted. These dynamic features and their characteristics are stored as dynamic nodes in the Neo4j graph database.

[0037] Step S230, construct association rules based on the interaction relationship between the static spatial features and the dynamic spatial features, and establish directed edges of the knowledge graph based on the association rules. Specifically, machine learning algorithms (such as Apriori, FP-Growth) are used to mine the association rules between static spatial features and dynamic spatial features. The interaction relationship between the features is analyzed (such as the relationship between traffic flow and road type, the relationship between air quality and building height). According to the mined association rules, directed edges of the knowledge graph are established. Each directed edge represents a causal relationship or interaction relationship, including the relationship type and weight.

[0038] For example, by analyzing the relationship between traffic flow and road type through the Apriori algorithm, it is found that "the traffic flow on main roads is usually higher than that on branch roads". This relationship is stored as a directed edge in the Neo4j graph database, and the weight of the edge is determined according to the support and confidence of the association rule.

[0039] Step S240: Based on the static nodes, the dynamic nodes, and the directed edges, construct a spatio-temporal causal knowledge graph. Specifically, use a graph database (such as Neo4j) to integrate the static nodes, dynamic nodes, and directed edges into a unified spatio-temporal causal knowledge graph. Optimize the structure of the graph through graph algorithms (such as path finding, community detection) to improve the query efficiency and analysis ability.

[0040] For example, integrate static nodes (buildings, roads), dynamic nodes (traffic flow, air quality), and directed edges (association rules) into the Neo4j graph database. Use Neo4j's Cypher query language for graph query and analysis, such as querying "which main roads have a traffic flow exceeding 1000 vehicles per hour during peak hours". By extracting static and dynamic spatial elements and their attributes and spatio-temporal change characteristics, a spatio-temporal causal knowledge graph containing static and dynamic information is constructed. This integration can provide a more comprehensive view of the urban operation status and provide richer data support for urban planning and management.

[0041] In a possible implementation, according to the interaction relationship between the static spatial elements and the dynamic spatial elements, construct association rules, and based on the association rules, establish the directed edges of the knowledge graph. Step S230 further includes step S231. If the spatial co-occurrence probability of any two static spatial elements is greater than a first preset threshold, generate a first dynamic association rule, and based on the first dynamic association rule, establish the first directed edge of the knowledge graph. Specifically, use spatial statistical methods (such as spatial autocorrelation analysis, spatial clustering analysis) to calculate the spatial co-occurrence probability of any two static spatial elements. For example, calculate the probability that two buildings appear in the same area, or the probability that two roads appear in the same traffic network. Set the first preset threshold (such as 0.8). If the spatial co-occurrence probability is greater than this threshold, it is considered that there is a significant spatial association between these two elements, and a first dynamic association rule is generated, indicating a strong spatial association between these two static spatial elements. In the knowledge graph, based on this association rule, establish the first directed edge, indicating the spatial association relationship from one element to another.

[0042] For example, assume that in urban planning, two buildings (A and B) often appear in the same area. Through spatial clustering analysis, the spatial co-occurrence probability between them is calculated to be 0.85, which is greater than the first preset threshold of 0.8. Therefore, a first dynamic association rule is generated, indicating a strong spatial association between building A and building B. In the knowledge graph, a first directed edge is established from building A to building B to represent this spatial association relationship.

[0043] Step S232: If the time-lag correlation between any two of the dynamic spatial elements is greater than the second preset threshold, generate a second dynamic association rule, and based on the second dynamic association rule, establish a second directed edge in the knowledge graph. Specifically, use time series analysis methods (such as cross-correlation analysis, Granger causality test) to calculate the time-lag correlation between any two dynamic spatial elements. For example, calculate the time-lag correlation between traffic flow and air quality index, and analyze the delayed impact of traffic flow changes on air quality. Set the second preset threshold (such as 0.7). If the time-lag correlation is greater than this threshold, it is considered that there is a significant time association between these two elements, and a second dynamic association rule is generated, indicating a strong time association between these two dynamic spatial elements. In the knowledge graph, based on this association rule, a second directed edge is established to represent the time association relationship from one element to another.

[0044] For example, assume that the time series analysis of traffic flow data and air quality data shows that an increase in traffic flow will cause the air quality index to decrease after 2 hours, and the time-lag correlation is 0.75, which is greater than the second preset threshold of 0.7. Therefore, a second dynamic association rule is generated, indicating a strong time association between traffic flow and air quality. In the knowledge graph, a second directed edge is established from traffic flow to air quality to represent this time association relationship.

[0045] Step S233: If the dependence correlation between any one of the static spatial elements and any one of the dynamic spatial elements is greater than the third preset threshold, generate a third dynamic association rule, and based on the third dynamic association rule, establish a third directed edge in the knowledge graph. Specifically, use statistical methods (such as Pearson correlation coefficient, mutual information) to calculate the dependence correlation between any one static spatial element and any one dynamic spatial element. For example, calculate the dependence correlation between building height and surrounding air quality, and analyze the impact of building height on air quality. Set the third preset threshold (such as 0.6). If the dependence correlation is greater than this threshold, it is considered that there is a significant dependence relationship between these two elements, and a third dynamic association rule is generated, indicating a strong dependence relationship between the static spatial element and the dynamic spatial element. In the knowledge graph, based on this association rule, a third directed edge is established to represent the dependence relationship from the static element to the dynamic element.

[0046] For example, assume that the analysis discovers a dependency relationship between the building height and the surrounding air quality, with a dependency correlation of 0.65, which is greater than the third preset threshold of 0.6. Therefore, a third dynamic association rule is generated, indicating that the building height has a significant impact on air quality. In the knowledge graph, a third directed edge is established from the building height to the air quality to represent this dependency relationship. By calculating the spatial co-occurrence probability, time-lag correlation, and dependency correlation respectively, the association rules between static and dynamic spatial elements can be mined more precisely. This refined mining method can more accurately reflect the complex relationships between elements.

[0047] Step S300: Perform multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer containing a time series confidence interval.

[0048] Specifically, use a GIS simulation tool (such as the simulation plug-in of ArcGIS Pro or QGIS) in combination with the spatio-temporal causal knowledge graph in the graph database to perform joint simulation on multi-source layer data and simulate the change process of spatial phenomena. Integrate the model results from different sources through model fusion techniques (such as multi-model integration and model calibration) to improve the simulation accuracy. Use a time series prediction model (such as ARIMA or LSTM neural network) to predict the simulation results and calculate the confidence interval. These models can handle the dynamic changes of time series data and provide an estimate of the uncertainty of the prediction. Train the model using historical data and evaluate the performance of the model through methods such as cross-validation to ensure the reliability of the prediction results. Visualize the prediction results and confidence intervals as a thematic layer and use a GIS visualization tool (such as ArcMap or QGIS) to generate a thematic map. Display the prediction results in an intuitive way on the map through layer rendering techniques (such as symbolization and graduated rendering) for easy understanding and use by users.

[0049] For example, combine the spatio-temporal causal relationship between meteorological data and environmental monitoring data and use an LSTM neural network to perform time series prediction on the air quality index. The prediction results show the change trend of the air quality index within the next 24 hours and calculate the 95% confidence interval. Generate these prediction results and confidence intervals as a thematic layer and display them in QGIS, where users can intuitively see the prediction results and the range of uncertainty through the map.

[0050] In a possible implementation, multi-layer joint simulation is performed according to the spatio-temporal causal knowledge graph to generate a predicted thematic layer including a time series confidence interval. Step S300 further includes step S310 of performing time discretization processing on the spatio-temporal causal knowledge graph to generate a time propagation sequence. Specifically, continuous time series data is discretized into multiple time steps (such as every hour, every minute). For example, the time of a day is discretized into 24 time steps. The time window (such as a sliding window) technology is used to divide the time series data into multiple time segments, and each segment corresponds to a time step. According to the discretized time steps, a time propagation sequence is generated, which represents the propagation path of the node state within each time step. The time propagation sequence can be represented as a time series graph, where the state of each node at each time step is connected by a directed edge.

[0051] For example, assume there is a spatio-temporal causal knowledge graph including buildings (static nodes) and air quality (dynamic nodes). The time of a day is discretized into 24 time steps (one step per hour). For each time step, the state change of the air quality node is recorded, and a time propagation sequence is generated. For example, the time propagation sequence can be expressed as: (t0, air quality 0) → (t1, air quality 1) → … → (t 23 , air quality 23 ).

[0052] Step S320, based on the time propagation sequence, starting from the states of each node at the current moment, perform state transfer simulation along the directed edges to obtain a first predicted thematic layer. Specifically, starting from the states of each node at the current moment, according to the causal relationship of the time propagation sequence and the directed edges, simulate the transfer of the node states. A state transition model (such as a Markov model) or a rule-based model (such as an association rule-based inference engine) is used for state transfer simulation. The result of the simulation is visualized as a first predicted thematic layer, which represents the predicted values of the node states within each time step.

[0053] For example, assume that the current air quality is "good". According to the time propagation sequence and the causal relationship (such as an increase in traffic flow leading to a decrease in air quality), simulate the change in air quality within the next 24 hours. Use a Markov model for state transfer simulation to generate a first predicted thematic layer. This layer shows the predicted values of air quality at each time step. For example, it is predicted that the air quality will decrease during the traffic peak hours.

[0054] Step S330: Based on the first predicted thematic layer, perform a simulation of adding random noise to the dynamic nodes to obtain a second predicted thematic layer. Specifically, on the basis of the first predicted thematic layer, add random noise to the states of the dynamic nodes to simulate the uncertainties in the real world. Use a random noise model (such as Gaussian noise, Poisson noise) to add noise to the states of the dynamic nodes. Update the simulation result after adding noise as the second predicted thematic layer, representing the prediction result considering uncertainties.

[0055] For example, assume that the first predicted thematic layer shows the change in air quality within the next 24 hours. To simulate uncertainties, add Gaussian noise to the predicted values of air quality. For example, if the predicted value is "good", it may become "good ± 5%" after adding noise, generating a second predicted thematic layer that shows the air quality prediction considering uncertainties.

[0056] Step S340: Add confidence intervals to the second predicted thematic layer to obtain a predicted thematic layer containing time series confidence intervals. Specifically, use statistical methods (such as the bootstrap method, Bayesian method) to calculate the confidence intervals of the states of the dynamic nodes at each time step. The confidence interval represents the range of uncertainty of the predicted value. Add the confidence intervals to the second predicted thematic layer to generate the final predicted thematic layer, representing the predicted values of the node states and their ranges of uncertainty at each time step.

[0057] For example, assume that the second predicted thematic layer shows the predicted values of air quality and their uncertainties within the next 24 hours. Use the bootstrap method to calculate the 95% confidence intervals of the air quality predicted values at each time step. For example, the predicted value is "good" and the confidence interval is "good ± 10%". Generate the final predicted thematic layer that shows the predicted values of air quality and their confidence intervals at each time step. Through time discretization processing and state transfer simulation, the dynamic changes of the node states in the spatio-temporal causal knowledge graph can be finely simulated. This refined simulation method can more accurately reflect the complex dynamic processes in the real world. By adding random noise to the dynamic nodes and calculating the confidence intervals, the uncertainties of the modeling prediction results can be effectively quantified and described. This uncertainty modeling method can provide more reliable prediction results to help decision-makers better evaluate risks.

[0058] In a possible implementation, a confidence interval is added to the second predicted thematic layer to obtain a predicted thematic layer containing a time series confidence interval. Step S340 further includes step S341 of obtaining the basic number of simulation runs N, where N ≥ 1000. Specifically, according to actual requirements and computing resources, the basic number of simulation runs N is set. N should be large enough to ensure the reliability of the statistical results. For example, in an environmental prediction system, N = 1000 simulation runs are set to ensure the stability and reliability of the statistical results.

[0059] Step S342, perform joint multi-layer simulation according to the basic number of simulation runs N to obtain N second predicted thematic layers. Specifically, use a GIS simulation tool (such as the simulation plug-in of ArcGIS Pro or QGIS) in combination with a spatio-temporal causal knowledge graph to perform joint multi-layer simulation. Each simulation generates a second predicted thematic layer, and a total of N simulation runs are performed to obtain N second predicted thematic layers.

[0060] For example, assume there are two dynamic layers of traffic flow and air quality. Use a GIS simulation tool to perform 1000 simulation runs. Each simulation generates a thematic layer containing traffic flow and air quality predictions, and finally 1000 second predicted thematic layers are obtained.

[0061] Step S343, adopt a Gaussian kernel density to calculate the probability distribution function for the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers. Specifically, use the Gaussian kernel density estimation method to calculate the probability distribution function of the spatio-temporal state values of each dynamic spatial element. Gaussian kernel density estimation is a non-parametric estimation method that can smoothly estimate the probability distribution of data.

[0062] For example, for each dynamic spatial element (such as the air quality index), extract its spatio-temporal state values from 1000 second predicted thematic layers, and use the scipy.stats.gaussian_kde function in Python to calculate its probability distribution function.

[0063] Step S344, for non-critical areas, take the 5% - 95% quantiles of the probability distribution function as the 90% confidence interval, and for critical areas, take the 10% - 90% quantiles of the probability distribution function as the 80% confidence interval. Specifically, calculate the quantiles according to the probability distribution function. For non-critical areas, calculate the 5% and 95% quantiles as the 90% confidence interval; for critical areas, calculate the 10% and 90% quantiles as the 80% confidence interval. Among them, the division of critical areas and non-critical areas is defined according to the requirements of the actual application scenario. For example, critical areas can be densely populated areas or environmentally sensitive areas.

[0064] For example, assume that a central area of a city is a key area and other areas are non-key areas. For the air quality index, calculate the 90% confidence interval (5% - 95% quantiles) for the non-key areas and the 80% confidence interval (10% - 90% quantiles) for the key areas.

[0065] Step S345: Calculate the mean of the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers to obtain a second mean predicted thematic layer. Specifically, calculate the mean of the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers to obtain the average predicted value of each element. Visualize these average predicted values as a second mean predicted thematic layer.

[0066] For example, for the air quality index, calculate the average value for each time step in 1000 second predicted thematic layers to generate a second mean predicted thematic layer. This layer shows the average predicted value of air quality for each time step.

[0067] Step S346: Add the confidence interval to the second mean predicted thematic layer to obtain a predicted thematic layer containing time series confidence intervals. Specifically, add the confidence interval calculated in step S344 to the second mean predicted thematic layer to generate a final predicted thematic layer. This layer not only contains the mean of the predicted values but also the confidence interval of the predicted values for each time step.

[0068] For example, based on the second mean predicted thematic layer, add the confidence interval of the air quality predicted value for each time step. For non-key areas, add a 90% confidence interval; for key areas, add an 80% confidence interval. The finally generated predicted thematic layer shows the average predicted value of air quality and its confidence interval for each time step. Through multiple simulation runs (N ≥ 1000) and statistical analysis, the generated predicted thematic layer can provide more reliable prediction results, and this method can effectively reduce the random error brought by a single simulation run.

[0069] Step S400: Calculate the dynamic time warping distance between the predicted thematic layer and the real-time monitoring layer, and generate a difference warning signal with spatial coordinate binding when the dynamic time warping distance is greater than the adaptive distance threshold.

[0070] Specifically, the dynamic time warping (DTW) algorithm is used to calculate the similarity between the time series of the predicted thematic layer and the real-time monitoring layer. The DTW algorithm can handle the stretching and deformation of time series and is suitable for the comparison of non-linear time series. The difference between the prediction result and the actual monitoring data is quantified by calculating the DTW distance. An adaptive distance threshold is dynamically calculated through statistical analysis (such as moving average, standard deviation) to determine whether the difference between time series is significant. The adaptive threshold can be automatically adjusted according to the statistical characteristics of the data to improve the robustness of the system. A dynamic update mechanism for the threshold is designed to ensure that the threshold can be adjusted according to the latest data to adapt to the changes in the data. When the DTW distance is greater than the adaptive threshold, GIS spatial analysis tools (such as the spatial analysis module of ArcGIS) are used to generate a difference warning signal bound to spatial coordinates. The specific spatial coordinate positions are marked on the GIS map to remind the user that there is a significant difference between the prediction and the actual monitoring in this area.

[0071] For example, the DTW algorithm is used to calculate the similarity between the predicted air quality index time series and the real-time monitored air quality index time series. If the DTW distance exceeds the adaptive threshold calculated by moving average and standard deviation (such as 1.5 times the standard deviation), a difference warning signal is generated on the GIS map, marking the specific spatial coordinate positions to remind the user that there is a significant difference between the prediction and the actual monitoring in this area.

[0072] In the embodiment of this application, technical means such as first accessing the data stream of multi-source dynamic layers with high-frequency updates in real time, then constructing a spatio-temporal causal knowledge graph with spatial elements as nodes and association rules as directed edges, then carrying out joint simulation of multiple layers based on this graph to generate a predicted thematic layer with a confidence interval of time series, and finally calculating the dynamic time warping distance between the predicted thematic layer and the real-time monitoring layer, and generating a difference warning signal bound to spatial coordinates when it is greater than the adaptive threshold are adopted, achieving the technical effect of improving the accuracy and reliability of prediction.

[0073] In a possible implementation manner, when calculating the dynamic time warping distance between the predicted thematic layer and the real-time monitoring layer, step S400 further includes step S410 of performing preprocessing on time series alignment for the predicted thematic layer and the real-time monitoring layer to obtain an aligned layer pair. Specifically, a time alignment algorithm (such as timestamp alignment, interpolation alignment) is used to align the time series of the predicted thematic layer and the real-time monitoring layer. Ensure that the time steps of the two layers are consistent for subsequent dynamic warping calculation.

[0074] For example, assume that the time step of the predicted thematic layer is once per hour, while the time step of the real-time monitoring layer is once every 30 minutes. Through interpolation alignment, the time step of the real-time monitoring layer is adjusted to once per hour to ensure that the time steps of the two layers are consistent. Finally, an aligned layer pair is obtained, and the time step of each layer is once per hour.

[0075] Step S420, perform multi-scale dynamic warping on the aligned layer pair to obtain a warped layer pair. Specifically, use the dynamic time warping algorithm to perform multi-scale dynamic warping on the aligned layer pair. The DTW algorithm can handle the stretching and deformation of time series and is suitable for the comparison of non-linear time series. Multi-scale dynamic warping can be performed at different time scales (such as hours, days, weeks) to capture changes at different time scales.

[0076] For example, for the aligned predicted thematic layer and real-time monitoring layer, use the DTW algorithm to perform multi-scale dynamic warping. At the hour scale, calculate the DTW distance between the two layers; at the day scale, aggregate the time series by day and calculate the DTW distance again. Finally, a warped layer pair is obtained, and each layer has undergone multi-scale dynamic warping.

[0077] Step S430, calculate the cosine similarity and root mean square error of the warped layer pair. Specifically, use the cosine similarity formula to calculate the similarity of the warped layer pair. Cosine similarity measures the angle between two vectors, and the value closer to 1 indicates greater similarity. Among them, the cosine similarity formula is: Cosine Among them, represents the dot product (inner product) of vector x and vector y. The dot product is the sum of the products of the corresponding elements of the two vectors. x i and y i respectively represent the elements of vector x and vector y at the i-th position, n represents the dimension of the vector (i.e., the number of elements), represents the modulus (length) of vector x. The modulus is the square root of the sum of the squares of all elements of the vector, represents the modulus (length) of vector y.

[0078] Use the root mean square error formula to calculate the error of the warped layer pair. The root mean square error measures the difference between the predicted value and the actual value, and the smaller the value, the smaller the error. Among them, the root mean square error formula is: Among them, m represents the length of the time series data, that is, the number of data points, p j represents the j-th data point in the predicted value sequence, q j represents the j-th data point in the actual value sequence.

[0079] For example, for the regularized prediction thematic layer and the real-time monitoring layer, calculate the cosine similarity and the root mean square error. Assume that the time series of the prediction layer is [10, 12, 15, 18], and the time series of the actual monitoring layer is [11, 13, 16, 19]. The calculated cosine similarity is 0.98, and the root mean square error is 1.

[0080] Step S440, weighted calculation of the cosine similarity and the root mean square error to obtain the dynamic time warping distance. Specifically, according to actual requirements, assign weights to the cosine similarity and the root mean square error, and perform weighted calculation of the dynamic time warping distance. The dynamic time warping distance formula is: DTW Distance = α·(1 - Cosine Similarity)+β·RMSE, where α and β are the weights of the cosine similarity and the root mean square error respectively, and α + β = 1.

[0081] For example, assume that the weights α = 0.6 and β = 0.4. According to the calculation results of step S430, the cosine similarity is 0.98, and the root mean square error is 1. Calculate the dynamic time warping distance as: DTW Distance = 0.6×(1 - 0.98)+0.4×1 = 0.412. By calculating the cosine similarity and the root mean square error, the similarity and error between the prediction thematic layer and the real-time monitoring layer can be evaluated in detail. The cosine similarity measures the similarity in direction, and the root mean square error measures the difference in values. The combination of the two provides a more comprehensive evaluation.

[0082] In a possible implementation, perform multi-scale dynamic warping on the aligned layer pair to obtain a regularized layer pair. Step S420 further includes step S421, perform coarse-grained alignment on the aligned layer pair to obtain a coarse-grained aligned layer pair. Specifically, use the dynamic time warping algorithm to perform coarse-grained alignment on the aligned layer pair. In the coarse-grained alignment stage, mainly focus on the overall shape and trend of the time series, and ignore local details. This can be achieved through a larger time window or a lower time resolution.

[0083] For example, assume there are two time series layers, the prediction layer c = [10, 12, 15, 18] and the actual monitoring layer d = [11, 13, 16, 19]. Use the DTW algorithm for coarse-grained alignment, aggregate the time series by a larger time window (such as every 2 time steps) to obtain a coarse-grained aligned layer pair. For example, the aggregated time series is: c ′ = [11, 16.5] and d ′ = [12, 17.5].

[0084] Step S422: Perform fine-grained optimization on the coarse-grained alignment layer pair to obtain a regularized layer pair. Specifically, on the basis of coarse-grained alignment, further optimize the alignment accuracy of the time series. This can be achieved through a smaller time window or a higher time resolution. Use local optimization algorithms (such as local adjustment, smoothing) to refine the coarse-grained alignment result to ensure that the time series is also aligned as much as possible in local details.

[0085] For example, on the basis of coarse-grained alignment, for the aggregated time series c ′ =[11, 16.5] and d ′ =[12, 17.5], perform fine-grained optimization. Use the local adjustment algorithm to fine-tune each time step to ensure that the time series is also aligned as much as possible in local details. For example, the optimized results are c″ = [11, 16.5] and d″ = [11.5, 17]. Coarse-grained alignment can quickly capture the overall similarity of the time series, while fine-grained optimization can further improve the alignment accuracy. This phased alignment method can effectively improve the accuracy and reliability of time series alignment.

[0086] In a possible implementation, when the dynamic time warping distance is greater than the adaptive distance threshold, a difference warning signal bound to the spatial coordinates is generated. Step S400 further includes step S450: Obtain the historical dynamic time warping distance within a preset time range, and use the 90th percentile of the historical dynamic time warping distance as the reference distance threshold. Specifically, obtain the dynamic time warping distance within a preset time range from the historical database. The preset time range can be the past month, three months, one year, etc., depending on the requirements of the application scenario. Calculate the 90th percentile of the historical dynamic time warping distance as the reference distance threshold. The 90th percentile means that 90% of the data points in the historical data have a dynamic time warping distance less than or equal to this value.

[0087] For example, assume that the dynamic time warping distance data within the past three months is obtained, and the 90th percentile of these data is calculated to be 1.5. Therefore, the reference distance threshold is set to 1.5.

[0088] Step S460: Calculate the quality score of the current data, and adaptively adjust the reference distance threshold according to the quality score to obtain the adaptive distance threshold. Specifically, calculate the quality score of the current data. The quality score is evaluated based on indicators such as data integrity, accuracy, and consistency. Adaptively adjust the reference distance threshold according to the quality score of the current data. If the data quality is high, appropriately lower the threshold; if the data quality is low, appropriately increase the threshold to ensure the rationality and reliability of the threshold.

[0089] For example, assume that the score of the current data quality is 0.8 (out of 1), indicating high data quality. According to the quality score, the reference distance threshold of 1.5 is adaptively adjusted to 1.2. If the quality score is low, such as 0.4, the reference distance threshold is adjusted to 1.8.

[0090] Step S470, when the dynamic time warping distance is greater than the adaptive distance threshold, generate multi-level difference warning signals bound to spatial coordinates according to the distance difference and mark the difference area on the map. Specifically, when the dynamic time warping distance is greater than the adaptive distance threshold, generate multi-level difference warning signals according to the distance difference. The multi-level warning signals are divided into different levels according to the size of the distance difference, such as low, medium, and high. Bind the difference warning signals to specific spatial coordinates to ensure that the difference area can be accurately located. Mark the difference area on the map, using different colors or symbols to represent different levels of warning signals, which is convenient for users to visually view the difference area.

[0091] For example, assume that the dynamic time warping distance is 1.6, which is greater than the adaptive distance threshold of 1.2. According to the distance difference (1.6 - 1.2 = 0.4), generate a medium-level difference warning signal. Bind this signal to specific spatial coordinates (such as longitude and latitude), and mark the medium-level warning area with a yellow mark on the map. If the distance difference is larger, such as 0.8, then generate a high-level warning signal and mark it with a red mark.

[0092] By calculating the 90th percentile of historical data as the reference distance threshold and performing adaptive adjustment according to the quality score of the current data, the rationality and reliability of the threshold can be ensured. This method can adapt to different data quality situations and improve the robustness of the system. Generating multi-level difference warning signals and performing spatial coordinate binding can more precisely reflect the difference between prediction and actual monitoring. The multi-level warning signals can provide different response measures according to the severity of the difference, enhancing the early warning ability of the system.

[0093] In the above text, reference is made to Figure 1 The GIS layer analysis method based on automatic map synthesis and parsing according to an embodiment of the present invention is described in detail. Next, reference will be made to Figure 2 Describe the GIS layer analysis system based on automatic map synthesis and parsing according to an embodiment of the present invention.

[0094] The GIS layer analysis system based on automatic map synthesis and parsing according to an embodiment of the present invention is used to solve the technical problems of insufficient prediction accuracy and reliability existing in the prior art, and achieve the technical effect of improving the prediction accuracy and reliability. The GIS layer analysis system based on automatic map synthesis and parsing includes: a multi-source dynamic layer data stream access module 10, a spatio-temporal causal knowledge graph construction module 20, a multi-layer joint simulation module 30, and a difference warning signal generation module 40.

[0095] The multi-source dynamic layer data stream access module 10 is used to access in real time a multi-source dynamic layer data stream updated at a frequency greater than a preset frequency; the spatio-temporal causal knowledge graph construction module 20 is used to construct a spatio-temporal causal knowledge graph with spatial elements as nodes and association rules as directed edges based on the multi-source dynamic layer data stream; the multi-layer joint simulation module 30 is used to perform multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval; the difference warning signal generation module 40 is used to calculate the dynamic time warping distance between the prediction thematic layer and the real-time monitoring layer, and generate a difference warning signal bound with spatial coordinates when the dynamic time warping distance is greater than an adaptive distance threshold.

[0096] Next, the specific configuration of the spatio-temporal causal knowledge graph construction module 20 will be described in detail. As described above, based on the multi-source dynamic layer data stream, a spatio-temporal causal knowledge graph is constructed with spatial elements as nodes and association rules as directed edges. The spatio-temporal causal knowledge graph construction module 20 may further include: a static node construction unit for extracting static spatial elements and corresponding attribute information in the multi-source dynamic layer data stream to construct static nodes of the knowledge graph; a dynamic node construction unit for extracting dynamic spatial elements and corresponding spatio-temporal change characteristics in the multi-source dynamic layer data stream to construct dynamic nodes of the knowledge graph; a directed edge establishment unit for constructing association rules according to the interaction relationship between the static spatial elements and the dynamic spatial elements, and establishing directed edges of the knowledge graph based on the association rules; a spatio-temporal causal knowledge graph construction unit for constructing a spatio-temporal causal knowledge graph based on the static nodes, the dynamic nodes and the directed edges.

[0097] Among them, according to the interaction relationship between the static spatial elements and the dynamic spatial elements, association rules are constructed, and directed edges of the knowledge graph are established based on the association rules. The directed edge establishment unit may further include: a first directed edge establishment subunit for generating a first dynamic association rule if the spatial co-occurrence probability of any two of the static spatial elements is greater than a first preset threshold, and establishing a first directed edge of the knowledge graph based on the first dynamic association rule; a second directed edge establishment subunit for generating a second dynamic association rule if the time lag correlation of any two of the dynamic spatial elements is greater than a second preset threshold, and establishing a second directed edge of the knowledge graph based on the second dynamic association rule; a third directed edge establishment subunit for generating a third dynamic association rule if the dependence correlation between any one of the static spatial elements and any one of the dynamic spatial elements is greater than a third preset threshold, and establishing a third directed edge of the knowledge graph based on the third dynamic association rule.

[0098] Next, the specific configuration of the multi-layer joint simulation module 30 will be described in detail. As described above, multi-layer joint simulation is performed according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval. The multi-layer joint simulation module 30 may further include: a time discretization processing unit for performing time discretization processing on the spatio-temporal causal knowledge graph to generate a time propagation sequence; a state transfer simulation unit for starting from the states of each node at the current moment based on the time propagation sequence and performing state transfer simulation along the directed edges to obtain a first prediction thematic layer; a random noise addition simulation unit for performing random noise addition simulation on the dynamic nodes based on the first prediction thematic layer to obtain a second prediction thematic layer; a confidence interval addition unit for adding a confidence interval to the second prediction thematic layer to obtain a prediction thematic layer including a time series confidence interval.

[0099] Among them, a confidence interval is added to the second predicted thematic layer to obtain a predicted thematic layer including a time series confidence interval. The confidence interval adding unit may further include: a simulation basic number obtaining subunit for obtaining the basic number N of simulations, where N≥1000; a multi-layer combined simulation subunit for performing multi-layer combined simulation according to the basic number N of simulations to obtain N second predicted thematic layers; a probability distribution function calculating subunit for calculating the probability distribution function of the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers by using a Gaussian kernel density; a confidence interval determining subunit for, for non-critical regions, taking the 5% - 95% quantiles of the probability distribution function as the 90% confidence interval, and for critical regions, taking the 10% - 90% quantiles of the probability distribution function as the 80% confidence interval; a mean calculating subunit for calculating the mean of the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers to obtain a second mean predicted thematic layer; and a confidence interval adding subunit for adding the confidence interval to the second mean predicted thematic layer to obtain a predicted thematic layer including a time series confidence interval.

[0100] Next, the specific configuration of the difference warning signal generation module 40 will be described in detail. As described above, the dynamic time warping distance between the predicted thematic layer and the real-time monitoring layer is calculated. The difference warning signal generation module 40 may further include: a time series alignment preprocessing unit for performing time series alignment preprocessing on the predicted thematic layer and the real-time monitoring layer to obtain an aligned layer pair; a multi-scale dynamic warping unit for performing multi-scale dynamic warping on the aligned layer pair to obtain a warped layer pair; a calculation unit for calculating the cosine similarity and root mean square error of the warped layer pair; and a weighting unit for calculating the cosine similarity and root mean square error by weighting to obtain the dynamic time warping distance.

[0101] Among them, when performing multi-scale dynamic warping on the aligned layer pair to obtain a warped layer pair, the multi-scale dynamic warping unit may further include: a coarse-grained alignment subunit for performing coarse-grained alignment on the aligned layer pair to obtain a coarse-grained aligned layer pair; and a fine-grained optimization subunit for performing fine-grained optimization on the coarse-grained aligned layer pair to obtain a warped layer pair.

[0102] Among them, when the dynamic time warping distance is greater than the adaptive distance threshold, a difference warning signal bound to spatial coordinates is generated. The difference warning signal generation module 40 may further include: a reference distance threshold acquisition unit for acquiring the historical dynamic time warping distance within a preset time range, and taking the 90th percentile of the historical dynamic time warping distance as the reference distance threshold; an adaptive distance threshold acquisition unit for calculating the quality score of the current data, and adaptively adjusting the reference distance threshold according to the quality score to obtain the adaptive distance threshold; a multi-level difference warning signal generation unit for generating a multi-level difference warning signal bound to spatial coordinates according to the distance difference and calibrating the difference area on the map when the dynamic time warping distance is greater than the adaptive distance threshold.

[0103] Next, the specific configuration of the multi-source dynamic layer data stream access module 10 will be described in detail. As described above, the multi-source dynamic layer data stream access module 10 may further include: a multi-source dynamic layer data stream preset unit for presetting a multi-source dynamic layer data stream, and the multi-source dynamic layer data stream includes: infrastructure sensor data, human activity density heat map, and environmental parameter time series change layer.

[0104] The GIS layer analysis system based on automatic map synthesis and parsing provided by the embodiments of the present invention can execute the GIS layer analysis method based on automatic map synthesis and parsing provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0105] Although this application makes various references to certain modules in the system according to the embodiments of this application, however, any number of different modules can be used and run on the user terminal and / or server. The included units and modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0106] The above specific implementation manners do not constitute a limitation to the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of this application shall be included within the protection scope of this application. In some cases, the actions or steps recorded in this application can be executed in a different order from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In certain embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

Claims

1. A GIS layer analysis method based on automatic map synthesis and parsing, characterized in that, The method includes: Real-time access to multi-source dynamic layer data streams updated at a frequency greater than a preset frequency; Based on the multi-source dynamic layer data streams, using spatial elements as nodes and association rules as directed edges, constructing a spatio-temporal causal knowledge graph; Performing multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval; Calculating the dynamic time warping distance between the prediction thematic layer and the real-time monitoring layer, and generating a difference warning signal with bound spatial coordinates when the dynamic time warping distance is greater than an adaptive distance threshold.

2. The GIS layer analysis method based on automatic map synthesis and parsing according to claim 1, characterized in that Based on the multi-source dynamic layer data streams, using spatial elements as nodes and association rules as directed edges, constructing a spatio-temporal causal knowledge graph, including: Extracting static spatial elements and corresponding attribute information in the multi-source dynamic layer data streams to construct static nodes of the knowledge graph; Extracting dynamic spatial elements and corresponding spatio-temporal change characteristics in the multi-source dynamic layer data streams to construct dynamic nodes of the knowledge graph; Constructing association rules according to the interaction relationship between the static spatial elements and the dynamic spatial elements, and establishing directed edges of the knowledge graph based on the association rules; Constructing a spatio-temporal causal knowledge graph based on the static nodes, the dynamic nodes and the directed edges.

3. The GIS layer analysis method based on automatic map synthesis and parsing according to claim 2, wherein Constructing association rules according to the interaction relationship between the static spatial elements and the dynamic spatial elements, and establishing directed edges of the knowledge graph based on the association rules, including: If the spatial co-occurrence probability of any two of the static spatial elements is greater than a first preset threshold, generating a first dynamic association rule and establishing a first directed edge of the knowledge graph based on the first dynamic association rule; If the time lag correlation of any two of the dynamic spatial elements is greater than a second preset threshold, generating a second dynamic association rule and establishing a second directed edge of the knowledge graph based on the second dynamic association rule; If the dependence correlation between any one of the static spatial elements and any one of the dynamic spatial elements is greater than a third preset threshold, generating a third dynamic association rule and establishing a third directed edge of the knowledge graph based on the third dynamic association rule.

4. The GIS layer analysis method based on automatic map synthesis and parsing according to claim 1, wherein, Performing multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a prediction thematic layer including a time series confidence interval, including: Performing time discretization processing on the spatio-temporal causal knowledge graph to generate a time propagation sequence; Based on the time propagation sequence, starting from the states of each node at the current moment, performing state transfer simulation along the directed edges to obtain a first prediction thematic layer; Based on the first prediction thematic layer, performing simulation of adding random noise to the dynamic nodes to obtain a second prediction thematic layer; Adding a confidence interval to the second prediction thematic layer to obtain a prediction thematic layer including a time series confidence interval.

5. The GIS layer analysis method based on automatic map synthesis and parsing according to claim 4, characterized in that Adding a confidence interval to the second prediction thematic layer to obtain a prediction thematic layer including a time series confidence interval, including: Obtaining the basic number of simulation times N, N≥1000; Performing multi-layer joint simulation according to the basic number of simulation times N to obtain N second prediction thematic layers; For the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers, calculate the probability distribution function using the Gaussian kernel density; For non-critical regions, take the 5% - 95% quantiles of the probability distribution function as the 90% confidence interval, and for critical regions, take the 10% - 90% quantiles of the probability distribution function as the 80% confidence interval; Calculate the mean value of the spatio-temporal state values of each dynamic spatial element in the N second predicted thematic layers to obtain the second mean predicted thematic layer; Add the confidence interval to the second mean predicted thematic layer to obtain a predicted thematic layer containing the time series confidence interval.

6. The GIS layer analysis method based on automatic map combination and parsing according to claim 1, characterized in that Calculate the dynamic time warping distance between the predicted thematic layer and the real-time monitoring layer, including: Perform preprocessing of time series alignment on the predicted thematic layer and the real-time monitoring layer to obtain an aligned layer pair; Perform multi-scale dynamic warping on the aligned layer pair to obtain a warped layer pair; Calculate the cosine similarity and root mean square error of the warped layer pair; Calculate the dynamic time warping distance by weighted calculation of the cosine similarity and root mean square error.

7. The GIS layer analysis method based on automatic map combination and parsing according to claim 6, characterized in that Perform multi-scale dynamic warping on the aligned layer pair to obtain a warped layer pair, including: Perform coarse-grained alignment on the aligned layer pair to obtain a coarse-grained aligned layer pair; Perform fine-grained optimization on the coarse-grained aligned layer pair to obtain a warped layer pair.

8. The GIS layer analysis method based on automatic map synthesis and parsing according to claim 1, characterized in that Generate a difference warning signal bound to spatial coordinates when the dynamic time warping distance is greater than the adaptive distance threshold, including: Obtain the historical dynamic time warping distance within a preset time range, and take the 90% quantile of the historical dynamic time warping distance as the benchmark distance threshold; Calculate the quality score of the current data, and adaptively adjust the benchmark distance threshold according to the quality score to obtain the adaptive distance threshold; When the dynamic time warping distance is greater than the adaptive distance threshold, generate a multi-level difference warning signal bound to spatial coordinates according to the distance difference and mark the difference area on the map.

9. The GIS layer analysis method based on automatic map synthesis and parsing according to claim 1, wherein The multi-source dynamic layer data stream includes: infrastructure sensor data, human activity density heat map, and environmental parameter time series change layer.

10. A GIS layer analysis system based on automatic map synthesis and parsing, characterized in that, The system is used to implement the GIS layer analysis method based on automatic map synthesis and parsing according to any one of claims 1-9. The system includes: A multi-source dynamic layer data stream access module for real-time accessing a multi-source dynamic layer data stream updated at a frequency greater than a preset frequency; A spatio-temporal causal knowledge graph construction module for constructing a spatio-temporal causal knowledge graph based on the multi-source dynamic layer data stream, with spatial elements as nodes and association rules as directed edges; A multi-layer joint simulation module for performing multi-layer joint simulation according to the spatio-temporal causal knowledge graph to generate a predicted thematic layer containing a time series confidence interval; A difference warning signal generation module for calculating the dynamic time warping distance between the predicted thematic layer and the real-time monitoring layer, and generating a difference warning signal bound to spatial coordinates when the dynamic time warping distance is greater than the adaptive distance threshold.

Images