Two-dimensional and three-dimensional integrated map display method and system

Abstract

The invention belongs to the technical field of information geography, and discloses a two-dimensional and three-dimensional integrated map display method and system. The method comprises the following steps: constructing an integrated map simulation model by using a three-dimensional reconstruction algorithm based on a cloud data center according to two-dimensional and three-dimensional geographic information data; according to the real-time monitoring data of the physical entity, using an OPC UA communication scheme to generate a data stream of a digital entity in the integrated map simulation model, and obtaining an integrated map digital twin model; and performing anomaly detection on the data stream to obtain a real-time anomaly detection result, and performing visual display on the data stream and the real-time anomaly detection result by using an integrated map digital twin model. The problems that in the prior art, deep integration of two-dimensional data and three-dimensional data is lacked, integration and display are not visual and efficient enough, a three-dimensional model can have large deviation, and the automation degree is low are solved.

Description

Technical Field

[0001] The present invention belongs to the field of information geography technology, and specifically relates to a two-dimensional and three-dimensional integrated map display method and system. Background Art

[0002] Geographic information data refers to the general term for all data used to represent the spatial distribution of the earth's surface. It contains the attribute information of various natural and artificial features on the earth's surface, as well as their spatial locations and relationships. Geographic information data is the core component of the geographic information system and is mainly used in many fields such as urban planning, resource management, environmental monitoring, disaster warning, transportation, market analysis, etc. Geographic information maps built based on geographic information data are used to display and analyze geographic information. In the geographic information system, geographic information maps are not only static visual representations, but also dynamic and interactive analysis tools. Geographic information maps can integrate multi-source data, such as remote sensing images, terrain data, demographics, etc., to provide a comprehensive view of spatial information, which can help users understand and interpret spatial data, conduct geographic analysis, and support decision making.

[0003] The existing geographic information map display has the following defects: 1) In existing technologies, the fusion of 2D and 3D geographic information data is usually relatively simple, lacking the integration of deep spatial relationships and attribute information. The conversion and synchronization between 2D data and 3D models are difficult, resulting in users being unable to obtain a complete information view when analyzing and making decisions. 2) Existing technologies often lack the ability to monitor and update the status of physical entities in real time, which is a major flaw for application scenarios that require real-time decision support. The integration and display of real-time data is usually not intuitive and efficient enough to meet the needs of rapid response in emergency situations; 3) In existing technologies, 3D reconstruction technology may be insufficient in accuracy, especially when dealing with complex terrain and building structures. The reconstructed 3D model may deviate greatly from the actual situation. The degree of automation in the 3D reconstruction process is not high, and often requires a lot of manual intervention, which increases the possibility of errors. Summary of the invention

[0004] In order to solve the problems in the prior art of lack of deep integration of two- and three-dimensional data, insufficient intuitiveness and efficiency in integration and display, large deviations in three-dimensional models and low degree of automation, the present invention aims to provide a two- and three-dimensional integrated map display method and system.

[0005] The technical solution adopted by the present invention is: A two-dimensional and three-dimensional integrated map display method includes the following steps: Based on the cloud data center, an integrated map simulation model is constructed using a 3D reconstruction algorithm based on 2D and 3D geographic information data; Based on the real-time monitoring data of the physical entity, the OPC UA communication solution is used to generate the data stream of the digital entity in the integrated map simulation model and obtain the integrated map digital twin model; Perform anomaly detection on the data stream to obtain real-time anomaly detection results, and use the integrated map digital twin model to visualize the data stream and real-time anomaly detection results.

[0006] Further, the two-dimensional and three-dimensional geographic information data includes two-dimensional geographic information data and three-dimensional geographic information data; Two-dimensional geographic information data includes two-dimensional spatial data, two-dimensional attribute data and two-dimensional metadata; Three-dimensional geographic information data includes three-dimensional spatial data, three-dimensional attribute data and three-dimensional metadata.

[0007] Furthermore, based on the cloud data center, according to the two-dimensional and three-dimensional geographic information data, a three-dimensional reconstruction algorithm is used to construct an integrated map simulation model, including the following steps: Based on the cloud data center, the two-dimensional and three-dimensional geographic information data of the target area are collected and analyzed to obtain the two-dimensional and three-dimensional geographic information data of the target area; According to the two-dimensional geographic information data, a three-dimensional reconstruction model based on a deep learning algorithm is used to perform three-dimensional reconstruction to obtain an initial integrated map simulation model; According to the three-dimensional geographic information data, the initial integrated map simulation model is modified to obtain the final integrated map simulation model.

[0008] Furthermore, the three-dimensional reconstruction model is constructed based on the CNN-Elman-GAN algorithm, and the three-dimensional reconstruction model includes a two-dimensional feature extraction module constructed based on the CNN algorithm, a target detection module constructed based on the Elman algorithm, and a three-dimensional reconstruction module constructed based on the GAN algorithm, which are connected in sequence.

[0009] Furthermore, according to the two-dimensional geographic information data, a three-dimensional reconstruction model constructed based on a deep learning algorithm is used to perform three-dimensional reconstruction to obtain an initial integrated map simulation model, including the following steps: Using a two-dimensional feature extraction module, extracting two-dimensional data features of the two-dimensional geographic information data, and inputting the two-dimensional data features into a target detection module; Use the target detection module to perform target detection based on the two-dimensional data features to obtain the physical target corresponding to the physical entity, extract the corresponding two-dimensional data features of the physical target, and input them into the three-dimensional reconstruction module; Using a 3D reconstruction module, 3D reconstruction is performed according to the 2D data features of the physical target to obtain a 3D model of the digital entity corresponding to the physical target; The physical target three-dimensional models and two-dimensional geographic information data of all physical targets are integrated to obtain an initial integrated map simulation model including several digital entities.

[0010] Furthermore, the initial integrated map simulation model is modified according to the three-dimensional geographic information data to obtain a final integrated map simulation model, including the following steps: Using a feature point extraction algorithm, extracting a number of first feature points of the three-dimensional geographic information data and a number of second feature points of the initial integrated map simulation model; Using a feature point matching algorithm, performing feature point matching on a plurality of first feature points and a plurality of second feature points to obtain a plurality of matching feature point pairs; According to a number of matching feature point pairs, the matching area of ​​the initial integrated map simulation model is corrected using three-dimensional geographic information data to obtain a final integrated map simulation model.

[0011] Furthermore, based on the real-time monitoring data of the physical entity, the OPC UA communication solution is used to generate the data stream of the digital entity in the integrated map simulation model, and the integrated map digital twin model is obtained, including the following steps: According to the physical target output by the 3D reconstruction model, collect the real-time monitoring data of the corresponding physical entity, and build the corresponding OPC UA information model based on the real-time monitoring data; Deploy an OPC UA server in the cloud data center and create a corresponding OPC UA instance in the address space of the OPC UA server according to the OPC UA information model; Transmit the real-time monitoring data to the OPC UA server in the cloud data center, write it to the corresponding OPC UA instance, and extract the real-time information data of the OPC UA instance; Convert real-time information data into data streams, input the data streams into the digital entities corresponding to the physical entities, and set up a visualization interface for the digital entities to obtain an integrated map digital twin model.

[0012] Furthermore, anomaly detection is performed on the data stream to obtain real-time anomaly detection results, and the data stream and real-time anomaly detection results are visualized using an integrated map digital twin model, including the following steps: Input the data stream of digital entities into the anomaly detection model built based on the deep learning algorithm, perform anomaly detection on the data stream, and obtain real-time anomaly detection results of digital entities; If the real-time anomaly detection result indicates that an anomaly exists, a corresponding real-time alarm signal is generated according to the real-time anomaly detection result; The data flow, real-time anomaly detection results and real-time alarm signals are displayed in the visualization interface of the digital entity corresponding to the integrated map digital twin model.

[0013] Furthermore, the anomaly detection model is constructed based on the GCN-RF algorithm, and the anomaly detection model includes a graph feature extraction module constructed based on the GCN algorithm and an anomaly detection module constructed based on the RF algorithm, which are connected in sequence.

[0014] A two-dimensional and three-dimensional integrated map display system is used to implement a two-dimensional and three-dimensional integrated map display method. The system includes a cloud data center and several data acquisition devices. The cloud data center is respectively connected to the several data acquisition devices for communication; The cloud data center is equipped with a three-dimensional reconstruction unit, a digital twin construction unit and a map display unit which are connected in sequence.

[0015] The beneficial effects of the present invention are: The present invention discloses a two-dimensional and three-dimensional integrated map display method and system. An integrated map simulation model is constructed by a three-dimensional reconstruction algorithm that combines two-dimensional and three-dimensional data fusion, which realizes the conversion from two-dimensional data to accurate three-dimensional model, effectively integrates the rich information of two-dimensional data and the stereoscopic perception of three-dimensional model, and provides a comprehensive spatial information view; the integrated map digital twin model realizes real-time digital twin monitoring and display, collects data of physical entities in real time through OPC UA information model, and integrates it into the digital twin model, realizing real-time synchronization between the physical world and the digital copy, while realizing intuitive and efficient integration and display, and greatly improving the efficiency of monitoring and decision support; when constructing the integrated map simulation model, a high-precision three-dimensional reconstruction and correction algorithm is introduced, and the initial three-dimensional model is accurately corrected using real three-dimensional data, ensuring the high consistency of the digital twin model with the actual physical entity, reducing the deviation from the actual situation, and improving the degree of automation, avoiding a large amount of manual intervention, and reducing the possibility of errors.

[0016] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of the two-dimensional and three-dimensional integrated map display method in the present invention.

[0018] Figure 2 It is a structural block diagram of the two-dimensional and three-dimensional integrated map display system in the present invention. DETAILED DESCRIPTION

[0019] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0020] Embodiment 1: like Figure 1 As shown, this embodiment provides a two-dimensional and three-dimensional integrated map display method, including the following steps: S1: Based on the cloud data center, according to the 2D and 3D geographic information data, using the 3D reconstruction algorithm, build an integrated map simulation model, including the following steps: S1-1: Based on the cloud data center, collect the two-dimensional and three-dimensional geographic information data of the target area, and analyze them to obtain the two-dimensional and three-dimensional geographic information data of the target area; Two or three-dimensional geographic information data includes two-dimensional geographic information data and three-dimensional geographic information data; Two-dimensional geographic information data is usually used to represent the surface information of the target area, including points, lines, surfaces and other elements on the map. It can be satellite images, aerial photos or maps. For example, the roads, administrative division boundaries, land use types and other aspects of the target area are all two-dimensional geographic information data. Two-dimensional geographic information data is usually represented on a plane without considering height or three-dimensional information. Three-dimensional geographic information data not only contains the information in the 3D geographic information data, but also adds the dimension in the vertical direction, that is, height information. Three-dimensional geographic information data more realistically represents the information of physical entities such as terrain, buildings, underground facilities, etc. in three dimensions. Two-dimensional geographic information data includes two-dimensional spatial data, two-dimensional attribute data and two-dimensional metadata; Two-dimensional spatial data describes the location and shape of surface features of the target area, usually in the form of points, lines, and surfaces. Two-dimensional attribute data provides descriptive information of two-dimensional spatial elements and is associated with two-dimensional spatial data. For example, the physical entity corresponding to a two-dimensional spatial data point may represent a building, and its two-dimensional attribute data may include the name, purpose, construction year, ownership, and other information of the building. Two-dimensional attribute data usually exists in a table form, corresponding to each element of the two-dimensional spatial data, and is associated with the spatial element through a unique identifier (such as ID). Two-dimensional metadata is data element information about two-dimensional geographic information data. In a geographic information system, two-dimensional metadata provides detailed information about spatial data sets, such as the source of two-dimensional geographic information data, creation date, data accuracy, data format, coordinate system, data collection method, data responsible unit, etc. Two-dimensional metadata is crucial for the management, sharing, and reuse of two-dimensional geographic information data, helping users understand and use two-dimensional geographic information data. Three-dimensional geographic information data includes three-dimensional spatial data, three-dimensional attribute data and three-dimensional metadata; In addition to the information of the two-dimensional spatial data, the three-dimensional spatial data also contains height or depth information, which can more accurately represent the three-dimensional features in the real world. The three-dimensional attribute data and the three-dimensional metadata are the same as the two-dimensional attribute data and the two-dimensional metadata, are related to the corresponding three-dimensional spatial data, and are used to understand and use the three-dimensional geographic information data; S1-2: Based on the two-dimensional geographic information data, a three-dimensional reconstruction model built based on a deep learning algorithm is used to perform three-dimensional reconstruction to obtain an initial integrated map simulation model; The 3D reconstruction model is constructed based on the CNN-Elman-GAN algorithm, and the 3D reconstruction model includes a 2D feature extraction module constructed based on the Convolutional Neural Networks (CNN) algorithm, a target detection module constructed based on the Elman algorithm, and a 3D reconstruction module constructed based on the Generative Adversarial Networks (GAN) algorithm, which are sequentially connected; The 2D feature extraction module is used to extract high-level features from 2D geographic information data. These features include edges, textures, shapes, etc., which can capture the details and structural information of 2D geographic information data, which is crucial for subsequent 3D reconstruction. The target detection module is used to detect and locate the target position and area of ​​the physical target in the two-dimensional geographic information data based on the two-dimensional data features. It can use time dynamic information for target detection, which helps to locate the physical target more accurately. The output will be used to guide the three-dimensional reconstruction process to ensure that the reconstructed three-dimensional model is aligned with the two-dimensional physical target in the two-dimensional geographic information data; The 3D reconstruction module consists of a generator and a discriminator. The task of the generator is to generate a 3D model using 2D features and target detection information, while the task of the discriminator is to distinguish between the generated 3D model and the real 3D model. Through this adversarial process, the generator can generate more and more realistic 3D models. The generator is usually a deep neural network that receives 2D features and target detection information as input and outputs a 3D model. The discriminator is another deep neural network that receives the 3D model as input and outputs a score indicating the authenticity of the model. The discriminator receives the 3D model output by the generator and tries to distinguish whether it is real or generated. Through multiple iterative training, the generator gradually learns to generate more realistic 3D models. According to the two-dimensional geographic information data, a three-dimensional reconstruction model based on a deep learning algorithm is used to perform three-dimensional reconstruction to obtain an initial integrated map simulation model, including the following steps: S1-2-1: using a two-dimensional feature extraction module to extract two-dimensional data features of the two-dimensional geographic information data, and inputting the two-dimensional data features into a target detection module; S1-2-2: Use the target detection module to perform target detection based on the two-dimensional data features to obtain the physical target corresponding to the physical entity, extract the corresponding two-dimensional data features of the physical target, and input them into the three-dimensional reconstruction module; S1-2-3: Using a 3D reconstruction module, perform 3D reconstruction according to the 2D data features of the physical target to obtain a 3D model of the digital entity corresponding to the physical target; S1-2-4: Integrate the physical target three-dimensional models and two-dimensional geographic information data of all physical targets to obtain an initial integrated map simulation model including several digital entities; S1-3: According to the three-dimensional geographic information data, the initial integrated map simulation model is corrected to obtain the final integrated map simulation model, including the following steps: S1-3-1: using a feature point extraction algorithm to extract a number of first feature points of the three-dimensional geographic information data and a number of second feature points of the initial integrated map simulation model; Feature point extraction algorithms include Scale-Invariant Feature Transform (SIFT), Speeded Up Robust Features (SURF), Oriented FAST and Rotated BRIEF (ORB), etc. S1-3-2: using a feature point matching algorithm, performing feature point matching on a plurality of first feature points and a plurality of second feature points to obtain a plurality of matching feature point pairs; Feature point matching algorithms include Brute-Force Matcher (BFM) algorithm, Fast Library for Approximate Nearest Neighbors (FLANN) matching algorithm, and Random Sample Consensus (RANSAC) algorithm. S1-3-3: According to a number of matching feature point pairs, using the three-dimensional geographic information data, the matching area of ​​the initial integrated map simulation model is corrected to obtain a final integrated map simulation model; S2: Based on the real-time monitoring data of the physical entity, the Object Linking and Embedding for Process Control Unified Architecture (OPCUA) communication solution is used to generate the data flow of the digital entity in the integrated map simulation model and obtain the integrated map digital twin model, including the following steps: S2-1: According to the physical target output by the 3D reconstruction model, collect the real-time monitoring data of the corresponding physical entity, and build the corresponding OPC UA information model according to the real-time monitoring data, including the following steps: S2-1-1: According to the physical target output by the three-dimensional reconstruction model, use a data acquisition device to collect real-time monitoring data of the corresponding physical entity, and pre-process the real-time monitoring data to obtain pre-processed real-time monitoring data; S2-1-2: Perform data analysis on the real-time monitoring data after preprocessing to obtain a number of physical parameters and physical parameter relationships, and set corresponding nodes, node attributes and node relationships according to the physical parameters and physical parameter relationships; Nodes are the building blocks in the OPC UA Information Model and include, but are not limited to: Object nodes: represent physical entities in the real world, such as terrain, buildings, underground facilities, etc. corresponding to physical targets; Variable nodes: represent data values, such as the monitored temperature, humidity, and pressure of terrain, buildings, and underground facilities; Method node: represents a function or operation that can be called; View Node: represents a subset of the OPC UA Information Model and can be used to organize nodes or provide a specific perspective; Node attributes are descriptive attributes of each node. These attributes define the characteristics of the node, including but not limited to: NodeId: unique identifier; NodeClass: the type of node (such as object, variable, method, etc.); BrowseName: the name used for browsing; DisplayName: the name used for display; Description: text describing the purpose of the node; References: a list of references pointing to other nodes; Value: For variable nodes, this attribute contains the current value of the variable; DataType: For variable nodes, this attribute defines the type of the value; Node relationship references are associations between nodes. They define the structure and relationship between nodes. References have a type and a direction. References allow building hierarchies and define how nodes relate to each other in the address space. S2-1-3: Construct the OPC UA information model of the physical entity based on the nodes, node attributes and node relationships, extract the model metadata of the OPC UA information model, which defines the structure of the information model, including objects, variables, methods, references, etc., and send it to the OPC UA server; S2-2: Deploy an OPC UA server in the cloud data center and create a corresponding OPC UA instance in the address space of the OPC UA server according to the OPC UA information model, including the following steps: S2-2-1: Deploy the OPC UA server in the cloud data center and collect model metadata of all OPC UA information models during initialization; S2-2-2: Create a corresponding OPC UA instance in the address space of the OPC UA server according to the model metadata; S2-3: Transmit the real-time monitoring data to the OPC UA server in the cloud data center, write it to the corresponding OPC UA instance, and extract the real-time information data of the OPC UA instance, including the following steps: S2-3-1: Use the data acquisition device to upload the real-time monitoring data to the OPC UA server transmitted to the cloud data center according to the OPC UA protocol; S2-3-2: Encapsulate the real-time node value and the real-time node attribute value in the real-time monitoring data to obtain corresponding real-time node encapsulation data; S2-3-3: Use the write service of the OPC UA protocol to write the real-time node encapsulation data to the node of the OPC UA instance; S2-3-4: When the preset data collection cycle arrives, according to the preset subscription service, the graph structure data of the OPC UA instance is extracted, including the real-time node value and real-time node attribute value of the node and the node relationship between the nodes, to obtain the real-time information data in the Extensible Markup Language (XML) format; S2-4: Convert the real-time information data into a data stream, input the data stream into the digital entity corresponding to the physical entity, and set a visualization interface for the digital entity to obtain an integrated map digital twin model; By building a unified OPC UA information model, standardized conversion between different data sources is achieved, the data processing process is simplified, and processing efficiency is improved. Through the data interaction processing process between the data acquisition device and the OPC UA server, it can quickly respond to the writing and query of real-time data to meet real-time requirements; S3: Perform anomaly detection on the data stream to obtain real-time anomaly detection results, and use the integrated map digital twin model to visualize the data stream and real-time anomaly detection results, including the following steps: S3-1: Input the data stream of the digital entity into the anomaly detection model built based on the deep learning algorithm, perform anomaly detection on the data stream, and obtain the real-time anomaly detection result of the digital entity; The anomaly detection model is built based on the Graph Convolutional Network (GCN)-Random Forest (RF) algorithm, and the anomaly detection model includes a graph feature extraction module built based on the GCN algorithm and an anomaly detection module built based on the RF algorithm, which are connected in sequence; The graph feature extraction module is used to extract the real-time graph features corresponding to the real-time information data in XML format. The real-time information data in XML format can be used to represent the graph structure information of the OPC UA instance. The graph is a mathematical structure composed of nodes (or vertices) and edges. The XML file can represent the nodes and edges in the graph by defining custom tags. In this embodiment, in order to accurately represent the graph structure information of the OPC UA instance, it is necessary to first restore the graph structure; The anomaly detection module filters the input graph features through the internal Classification And Regression Tree (CART), which can process a large number of features and select the most stable and discriminative graph feature components, and perform anomaly detection based on the graph feature components; The data stream of the digital entity is input into the anomaly detection model built based on the deep learning algorithm, and anomaly detection is performed on the data stream to obtain the real-time anomaly detection result of the digital entity, including the following steps: S3-1-1: Perform graph structure restoration based on real-time information data to obtain the corresponding real-time OPC UA information graph, and input the real-time OPC UA information graph into the graph feature extraction module of the anomaly detection model; S3-1-2: using the graph feature extraction module, extracting the real-time graph features of the real-time OPC UA information graph, and inputting the real-time graph features into the anomaly detection module of the anomaly detection model; S3-1-3: using an anomaly detection module, extracting several real-time key graph feature components of the real-time graph feature, and performing anomaly detection based on the several real-time key graph feature components to obtain a real-time anomaly detection result of the digital entity, including the following steps: S3-1-3-1: Use the RF structure trained in the anomaly detection module to extract the feature contribution of M real-time candidate graph feature components in the real-time graph feature, where M is the number of key graph feature components; The formula is:

[0021] In the formula, For the Feature contribution of feature components of real-time candidate graph; For the Real-time candidate graph feature components in random forest The feature contribution of each tree; is the CART tree indicator; It is an indicator of the characteristic component of the real-time candidate graph; is the total number of CARTs;

[0022] In the formula, CART tree node for random forest m ,node and nodes The Gini index of CART tree node m Medium Category The proportion of is the total number of categories; m , , is the node indicator; is the category indicator; S3-1-3-2: normalize the feature contributions of the M real-time candidate graph feature components to obtain a number of corresponding normalized feature contributions; The formula is:

[0023] In the formula, It is the contribution of the feature after normalization; J is the total number of feature components of the real-time candidate graph; S3-1-3-3: Generate feature selection standard values ​​for several real-time candidate graph feature components based on the normalized feature contribution; The formula is:

[0024] In the formula, For the Feature selection criteria values ​​for feature components of real-time candidate graphs; For the Normalized feature contribution of feature components of real-time candidate graph; It is an indicator of the characteristic component of the real-time candidate graph; S3-1-3-4: according to the feature selection standard value, the real-time candidate graph feature components are sorted in descending order, and the first M real-time candidate graph feature components are selected as the real-time key graph feature components to obtain M real-time key graph feature components; S3-1-3-5: Perform anomaly detection based on several real-time key graph feature components to obtain real-time anomaly detection results of digital entities; S3-2: If the real-time anomaly detection result indicates that an anomaly exists, a corresponding real-time alarm signal is generated according to the real-time anomaly detection result; S3-3: Display the data stream, real-time anomaly detection results, and real-time alarm signals in the visualization interface of the digital entity corresponding to the integrated map digital twin model; The anomaly detection model can effectively identify anomalies in data streams and visualize them through an integrated map digital twin model, providing strong technical support for real-time monitoring and early warning.

[0025] Embodiment 2: like Figure 2 As shown, this embodiment provides a two-dimensional and three-dimensional integrated map display system for implementing a two-dimensional and three-dimensional integrated map display method. The system includes a cloud data center and several data acquisition devices. The cloud data center is respectively connected to the several data acquisition devices for communication; A data collection device, used to collect real-time monitoring data of the physical entity, and send the real-time monitoring data of the physical entity to a cloud data center according to the OPC UA protocol; The cloud data center is equipped with a three-dimensional reconstruction unit, a digital twin construction unit, and a map display unit that are connected in sequence; A three-dimensional reconstruction unit is used to construct an integrated map simulation model based on two-dimensional and three-dimensional geographic information data using a three-dimensional reconstruction algorithm; A digital twin construction unit is used to generate data streams of digital entities in an integrated map simulation model based on real-time monitoring data of physical entities using an OPC UA communication solution, and obtain an integrated map digital twin model; The map display unit is used to perform anomaly detection on the data stream, obtain real-time anomaly detection results, and use the integrated map digital twin model to visualize the data stream and real-time anomaly detection results.

[0026] The present invention discloses a two-dimensional and three-dimensional integrated map display method and system. An integrated map simulation model is constructed by a three-dimensional reconstruction algorithm that combines two-dimensional and three-dimensional data fusion, which realizes the conversion from two-dimensional data to accurate three-dimensional model, effectively integrates the rich information of two-dimensional data and the stereoscopic perception of three-dimensional model, and provides a comprehensive spatial information view; the integrated map digital twin model realizes real-time digital twin monitoring and display, collects data of physical entities in real time through OPC UA information model, and integrates it into the digital twin model, realizing real-time synchronization between the physical world and the digital copy, while realizing intuitive and efficient integration and display, and greatly improving the efficiency of monitoring and decision support; when constructing the integrated map simulation model, a high-precision three-dimensional reconstruction and correction algorithm is introduced, and the initial three-dimensional model is accurately corrected using real three-dimensional data, ensuring the high consistency of the digital twin model with the actual physical entity, reducing the deviation from the actual situation, and improving the degree of automation, avoiding a large amount of manual intervention, and reducing the possibility of errors.

[0027] The present invention is not limited to the above optional implementations, and anyone can derive other various forms of products under the enlightenment of the present invention. The above specific implementations should not be understood as limiting the scope of protection of the present invention. The scope of protection of the present invention should be based on the definition in the claims, and the description can be used to interpret the claims.

Claims

1. A two-dimensional and three-dimensional integrated map display method, characterized in that: The steps include: Based on the cloud data center, an integrated map simulation model is constructed using a 3D reconstruction algorithm based on 2D and 3D geographic information data; Based on the real-time monitoring data of the physical entity, the OPC UA communication solution is used to generate the data stream of the digital entity in the integrated map simulation model and obtain the integrated map digital twin model; Perform anomaly detection on the data stream to obtain real-time anomaly detection results, and use the integrated map digital twin model to visualize the data stream and real-time anomaly detection results.

2. A two-dimensional and three-dimensional integrated map display method according to claim 1, characterized in that: The two-dimensional and three-dimensional geographic information data includes two-dimensional geographic information data and three-dimensional geographic information data; The two-dimensional geographic information data includes two-dimensional spatial data, two-dimensional attribute data and two-dimensional metadata; The three-dimensional geographic information data includes three-dimensional space data, three-dimensional attribute data and three-dimensional metadata.

3. The method for displaying a two-dimensional and three-dimensional integrated map according to claim 2, characterized in that: Based on the cloud data center, according to the 2D and 3D geographic information data, using the 3D reconstruction algorithm, an integrated map simulation model is constructed, including the following steps: Based on the cloud data center, the two-dimensional and three-dimensional geographic information data of the target area are collected and analyzed to obtain the two-dimensional and three-dimensional geographic information data of the target area; According to the two-dimensional geographic information data, a three-dimensional reconstruction model based on a deep learning algorithm is used to perform three-dimensional reconstruction to obtain an initial integrated map simulation model; According to the three-dimensional geographic information data, the initial integrated map simulation model is modified to obtain the final integrated map simulation model.

4. A two-dimensional and three-dimensional integrated map display method according to claim 3, characterized in that: The three-dimensional reconstruction model is constructed based on the CNN-Elman-GAN algorithm, and the three-dimensional reconstruction model includes a two-dimensional feature extraction module constructed based on the CNN algorithm, a target detection module constructed based on the Elman algorithm, and a three-dimensional reconstruction module constructed based on the GAN algorithm, which are connected in sequence.

5. A two-dimensional and three-dimensional integrated map display method according to claim 4, characterized in that: According to the two-dimensional geographic information data, a three-dimensional reconstruction model based on a deep learning algorithm is used to perform three-dimensional reconstruction to obtain an initial integrated map simulation model, including the following steps: Using a two-dimensional feature extraction module, extracting two-dimensional data features of the two-dimensional geographic information data, and inputting the two-dimensional data features into a target detection module; Use the target detection module to perform target detection based on the two-dimensional data features to obtain the physical target corresponding to the physical entity, extract the corresponding two-dimensional data features of the physical target, and input them into the three-dimensional reconstruction module; Using a 3D reconstruction module, 3D reconstruction is performed according to the 2D data features of the physical target to obtain a 3D model of the digital entity corresponding to the physical target; The physical target three-dimensional models and two-dimensional geographic information data of all physical targets are integrated to obtain an initial integrated map simulation model including several digital entities.

6. The method for displaying a 2D and 3D integrated map according to claim 3, characterized in that: According to the three-dimensional geographic information data, the initial integrated map simulation model is corrected to obtain the final integrated map simulation model, including the following steps: Using a feature point extraction algorithm, extracting a number of first feature points of the three-dimensional geographic information data and a number of second feature points of the initial integrated map simulation model; Using a feature point matching algorithm, performing feature point matching on a plurality of first feature points and a plurality of second feature points to obtain a plurality of matching feature point pairs; According to a number of matching feature point pairs, the matching area of ​​the initial integrated map simulation model is corrected using three-dimensional geographic information data to obtain a final integrated map simulation model.

7. The method for displaying a 2D and 3D integrated map according to claim 5, characterized in that: Based on the real-time monitoring data of the physical entity, the OPC UA communication solution is used to generate the data stream of the digital entity in the integrated map simulation model, and the integrated map digital twin model is obtained, including the following steps: According to the physical target output by the 3D reconstruction model, collect the real-time monitoring data of the corresponding physical entity, and build the corresponding OPC UA information model based on the real-time monitoring data; Deploy an OPC UA server in the cloud data center and create a corresponding OPC UA instance in the address space of the OPC UA server according to the OPC UA information model; Transmit the real-time monitoring data to the OPC UA server in the cloud data center, write it to the corresponding OPC UA instance, and extract the real-time information data of the OPC UA instance; Convert real-time information data into data streams, input the data streams into the digital entities corresponding to the physical entities, set up a visualization interface for the digital entities, and obtain an integrated map digital twin model.

8. The method for displaying a 2D and 3D integrated map according to claim 7, characterized in that: Perform anomaly detection on the data stream to obtain real-time anomaly detection results, and use the integrated map digital twin model to visualize the data stream and real-time anomaly detection results, including the following steps: Input the data stream of the digital entity into the anomaly detection model built based on the deep learning algorithm, perform anomaly detection on the data stream, and obtain real-time anomaly detection results of the digital entity; If the real-time anomaly detection result indicates that an anomaly exists, a corresponding real-time alarm signal is generated according to the real-time anomaly detection result; The data flow, real-time anomaly detection results and real-time alarm signals are displayed in the visualization interface of the digital entity corresponding to the integrated map digital twin model.

9. The method for displaying a 2D and 3D integrated map according to claim 8, characterized in that: The anomaly detection model is constructed based on the GCN-RF algorithm, and the anomaly detection model includes a graph feature extraction module constructed based on the GCN algorithm and an anomaly detection module constructed based on the RF algorithm, which are connected in sequence.

10. A 2D and 3D integrated map display system, used to implement the 2D and 3D integrated map display method according to any one of claims 1 to 8, characterized in that: The system includes a cloud data center and several data acquisition devices, and the cloud data center is respectively connected to the several data acquisition devices for communication; The cloud data center is equipped with a three-dimensional reconstruction unit, a digital twin construction unit and a map display unit which are connected in sequence.

Images