Vector geographic data efficient visualization method and system based on multi-scale optimization and containerization architecture

Abstract

The invention provides a vector geographic data efficient visualization method and system based on multi-scale optimization and containerization architecture, and belongs to the field of spatial big data visualization and cloud native GIS application. According to the method, the filling rate of the enclosing rectangles and the number of the vertexes are calculated for the original geographic element data, so that automatic classification of regular elements and irregular elements is realized; and respectively constructing two sets of multi-scale expression systems, and storing four types of multi-scale data tables in a spatial database. A containerized micro-service architecture is adopted, and dependence management and cooperative operation are automatically completed through a container arrangement tool. The vector slicing service dynamically selects a corresponding scale expression according to the scaling level. Through cooperative work of a database trigger, a monitoring service and a cache service, the system can calculate an affected slice number range and execute accurate cache deletion when element attributes are updated, and real-time and on-demand slice updating is realized. The method has the beneficial effects that the visualization performance and updating efficiency of massive vector elements are improved.

Description

Technical Field

[0001] This invention relates to the field of spatial big data visualization and cloud-native containerized architecture, and in particular to a method and system for efficient visualization of vector geographic data based on multi-scale optimization and containerized architecture. Background Technology

[0002] In the construction and operation of projects such as the Internet of Things, smart energy, and smart cities, the number and level of detail of spatial elements that Geographic Information Systems (GIS) need to process are constantly increasing. For example, a large photovoltaic power station often contains hundreds of thousands or even millions of photovoltaic panel units, and the number of irregular polygonal elements such as building outlines and land boundaries within a city can reach millions. As the business side's requirements for refined operation and maintenance monitoring, spatial analysis, and real-time display continue to increase, WebGIS needs to perform high-frequency zooming, panning, and interactive operations on massive fine-grained vector elements on the browser side, which brings significant performance and scalability pressure to the existing map service architecture. Traditional WebGIS visualization mostly adopts WMS / WMTS raster tile services published by map servers such as GeoServer. This type of solution uses server-rendered map images as the main carrier: whenever the user zooms in, zooms out, or pans in the browser, the front end needs to send a new tile request to the server, which then renders and returns the corresponding raster image based on the current view range. This model suffers from several typical problems: First, server-side rendering pressure is high, with CPU and I / O loads rising sharply in scenarios with massive amounts of finely layered elements. Second, the large size of raster images leads to massive data transmission over the network, resulting in slow client loading and a poor user experience. Third, raster images are essentially "fixed," making it difficult for clients to dynamically switch feature-level styles, perform high-performance attribute queries, and conduct fine-grained interactive analysis, thus failing to meet the needs of real-time operation and maintenance monitoring and business linkage. To improve performance, rendering methods based on vector tiles (such as MVT) have emerged. Existing technologies, such as CN111353007A, propose using encoding matrices to achieve feature picking, improving interaction efficiency, but they do not solve the problems of multi-scale representation of regular and irregular features and data redundancy, nor do they have a hierarchical optimization strategy for large-scale data. Meanwhile, GIS systems are evolving towards cloud computing and containerized architectures. CN108009001A uses Docker to achieve elastic deployment of GIS services, improving operation and maintenance efficiency, but it does not address multi-scale optimization of geographic features or precise control of caching triggered by data updates. The inventors discovered that existing solutions either focus on vector tile rendering and picking or on system deployment architecture, lacking an end-to-end optimization mechanism from data preprocessing to client-side rendering. Faced with complex scenarios where regular and irregular elements coexist, there is still a lack of a high-performance vector data visualization method and system that simultaneously possesses intelligent classification, multi-scale representation, containerized collaboration, and precise cache update capabilities. Therefore, there is an urgent need to provide a full-link technical solution from geographic feature input, intelligent preprocessing, multi-scale optimization, to containerized vector tile publishing and collaborative cache management, to achieve high-performance, highly scalable, and real-time updated WebGIS vector data visualization in scenarios with massive amounts of small, regular / irregular geographic features. The descriptions in the background art are for illustrative purposes only and, except for explicitly cited documents or patents, do not constitute an admission of prior art. Summary of the Invention

[0003] The purpose of this invention is to address the problems of inefficient rendering, data redundancy, coarse-grained cache failure, and insufficient system scalability in the visualization of massive fine-grained geographic features. This is achieved by constructing a complete data processing chain from intelligent feature classification, multi-scale geometric optimization, dynamic loading of vector tiles, to containerized microservice collaborative operation. Combined with a feature geometric feature discrimination model, a multi-scale representation selection model, and a precise cache update mechanism based on spatial range calculation, this forms an intelligent visualization solution that enables automatic feature type identification, on-demand selection of optimized representations, targeted failure of tile cache, and stable high-concurrency operation across the entire chain. This significantly improves the rendering performance, real-time performance, and deployment scalability of large-scale regular and irregular features.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by this invention is: an efficient visualization method for vector geographic data based on multi-scale optimization and containerized architecture, comprising the following steps: S1. Data Input and Classification: Calculate the bounding rectangle fill rate and vertex number of the input raw geographic feature data, and classify the features into regular features and irregular features according to preset rules; S2, Intelligent Multi-level Optimization: Generates center point representation, simplified surface representation, and detailed surface representation for irregular elements; generates center point representation, grid aggregation unit representation, and independent element representation for regular elements; S3. Multi-scale data storage: The multi-scale representations obtained in step S2 are stored in multi-scale data tables, including point-level tables, simplified-level tables, grid aggregation tables, and detailed-level tables. S4. Containerized microservice deployment: The data preprocessing container, PostGIS spatial database container and vector tile container are launched sequentially through the container orchestration tool, enabling the tile module to generate vector tiles based on multi-scale optimized geographic feature data; S5. Hierarchical Slice Generation: The vector slice container dynamically selects the corresponding multi-scale data table according to the current scaling level; low scaling level uses center point or aggregate expression, medium scaling level uses simplified surface or mesh aggregate surface, and high scaling level uses detailed and complex geometry. S6. Collaborative Cache Management: When a feature attribute changes, the database trigger sends a message including the feature identifier and spatial information to the listening service. The listening service calculates the range of tile numbers covered by the feature and its grid cell at each scaling level. The cache service performs precise cache deletion based on the tile number.

[0005] In the preferred embodiment, the simplified surface representation of irregular elements in step S2 adopts the topology-preserving Douglas-Puk algorithm.

[0006] In the preferred embodiment, the grid aggregation expression of regular features in step S2 is generated by dividing the spatial region into regular grids and calculating the number of features within the grids.

[0007] In the preferred embodiment, in step S5, when the zoom level is higher than a preset threshold, a detailed surface or independent feature is loaded; when the zoom level is in the middle range, a simplified surface or mesh aggregation unit is loaded; and when the zoom level is lower, a center point representation is loaded.

[0008] In the preferred embodiment, the slice number calculation in step S6 includes determining the slice number covered by the updated feature at each scaling level based on the bounding rectangle of the updated feature; for features with grid aggregation expression, the slice number at the corresponding scaling level is further determined based on the geometric range of the grid cell to which it belongs.

[0009] In the preferred embodiment, in step S6, the cache service only deletes the cache key value corresponding to the slice number obtained in step S6, without deleting other unaffected caches.

[0010] This invention also provides a high-efficiency visualization system for vector geographic data based on multi-scale optimization and containerized architecture, applied to the methods described above. The system includes a data input and intelligent preprocessing module for receiving raw geographic feature data, classifying the raw features, and generating differentiated multi-scale representations; a containerized microservice deployment module, including a data preprocessing container, a PostGIS spatial database container, a vector tile container, a Redis cache container, and a front-end service container, which automatically starts and works collaboratively according to preset dependencies using a container orchestration tool; a client and rendering module for loading corresponding vector tiles according to the current view scale and visualizing different representation levels of regular and irregular features; and a collaborative cache management module, including database triggers, a listening service, and a cache service, for performing precise cache clearing when feature data is updated. This system achieves high-performance collaboration across the entire chain of data optimization, tile generation, cache updating, and rendering display through containerized deployment.

[0011] In the preferred embodiment, the data input and intelligent preprocessing module includes a feature classification unit, an irregular feature multi-scale optimization unit, and a regular feature multi-scale optimization unit.

[0012] In the preferred embodiment, the containerized microservice deployment module defines the dependencies between containers through Docker Compose or Kubernetes, so that the PostGIS container starts after the data preprocessing container completes its optimization tasks, and the vector tile container runs after the PostGIS container starts.

[0013] In the preferred embodiment, the collaborative cache management module triggers the vector tile container to regenerate the tile containing the updated feature based on the tile number calculated by the listening service.

[0014] Beneficial effects: (1) The scheme of centrally and intelligently processing the data expression form and vector tile generation strategy based on the regularity characteristics of geographic elements and the view zoom level can greatly improve the rendering efficiency, data update efficiency and system scalability of massive vector geographic elements in WebGIS scenarios.

[0015] (2) Construct two sets of multi-scale data expression pipelines: “point-grid aggregation-independent features” and “point-simplified surface-detailed surface”, so that the system can select the most suitable expression form as needed at different scaling levels, significantly reduce data redundancy in small-scale views, improve the geometric expression accuracy in large-scale views, and achieve a dynamic balance between performance and expression quality.

[0016] (3) By decoupling and deploying data preprocessing, PostGIS spatial database, Tegola vector tiling service, Redis cache and front-end rendering service through containerized microservice architecture, and automatically starting them according to the dependencies between containers, the automated collaboration from data source to front-end rendering is realized.

[0017] (4) By combining the collaborative mechanism of database triggers, listening services and caching services, the system can capture change events in real time when feature attributes change. By calculating the tile numbers covered by the changed feature itself and its grid at all zoom levels, the system obtains the set of tile caches that need to be cleared, performs precise cache updates, and ensures that the front end always loads the latest data, thus avoiding performance waste caused by global cache failure. Attached Figure Description

[0018] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a diagram of the multi-scale optimization strategy architecture of the present invention; Figure 2 This is the system architecture diagram of the present invention. Detailed Implementation

[0019] Example 1 like Figure 1 As shown, a method for efficient visualization of vector geographic data based on multi-scale optimization and containerized architecture is applied to a smart energy scenario. This method utilizes 800,000 geographic feature data points, including photovoltaic panels and buildings, uniformly stored in the raw feature data table `raw_features`. The method includes the following steps: S1. Perform classification on the input geographic feature data: First, all feature records are loaded from the raw_features table. Each feature contains a unique ID (id), a geometry field (geom), a feature type (feature_type), and relevant business attribute fields. Geometric indices are calculated to distinguish between features with regular layouts (e.g., photovoltaic panels) and features with complex shapes and irregular distributions (e.g., buildings). This process includes the following steps: S101. Calculate the circumscribed rectangle: (1); Where bbox represents the bounding rectangle geometry of the current feature, and ST_Envelope() is the spatial calculation function that generates the minimum bounding rectangle in the spatial database; S102. Calculate the fill rate: (2); Where fill_ratio represents the fill ratio, and ST_Area() is a function for calculating the area in the spatial database; the closer the fill ratio is to 1, the closer the shape of the feature is to a regular rectangular structure. S103, Calculate the number of vertices: (3); Where n_points is the number of vertices of the feature geometry, and ST_NPoints() is a function in the spatial database used to count the number of geometric points; S104. Classify according to classification rules: When fill_ratio > 0.85 and n_points = 5, the raw feature data is classified as regular features; otherwise, it is classified as irregular features. That is, when the area of ​​a feature matches the area and height of its bounding rectangle and the number of vertices matches the rectangle's structure, the feature can be identified as a regular feature. Other features with complex shapes or multiple polylines on their boundaries are classified as irregular features. The classification result is written back to the feature_type field of the raw_features table. S2. Perform multi-scale optimization based on feature type: Based on the classification results of step S1, the features enter two different data optimization processes: a three-layer representation system for irregular features and a three-layer representation system for regular features. The goal of this step is to significantly reduce data redundancy at low scaling levels while maintaining sufficient geometric detail at high scaling levels, thereby improving rendering efficiency. This includes the following steps: S201. Multi-scale optimization of irregular features: Irregular features typically refer to building outlines, complex polygonal plots, etc. The following three representations are generated for such features: Point-level expression: (4); Here, point_geom represents the calculated geometric center point position, and ST_Centroid is a function for finding the center point of a polygon in the spatial database; the center point is used to replace complex polygons when the scaling level is low, which can significantly reduce the amount of data.

[0020] Simplified surface representation: The outline is simplified using the Douglas-Puk algorithm, and the geometry is simplified using the following formula: (5); Here, simple_geom represents the simplified polygon, and ST_SimplifyPreserveTopology is a geometric simplification function that preserves the topology within the spatial database, with a simplification tolerance of 0.5 meters. When the simplified simple_geom remains geometrically valid, the system saves the simplification result. If the target geometry loses its topological relationships or its structure is destroyed, the original geometry is used to ensure that map symbols are displayed correctly.

[0021] Detailed surface description: The detailed representation maintains the original geometric accuracy and is used to directly display the complete outline at high zoom levels.

[0022] S202, Multi-scale optimization of rule elements: Regular elements typically refer to photovoltaic panel arrays, equipment ledgers with regular shapes, etc., and the following three forms of expression are generated for such elements: Point-level expression: (6); Here, panel_point represents the center point coordinates of the feature at the low zoom level.

[0023] Mesh aggregation expression: An equidistant, regular grid is constructed for the photovoltaic power station area. In this embodiment, the grid side length is fixed at 100 meters to present an overview of the area at a medium zoom level. The grid is generated using: (7); Where grid_geom is the generated grid polygon, and ST_MakeEnvelope is a function in the spatial database used to generate rectangular grids; the number of photovoltaic panels and the number of faults in each grid are counted, and the grid aggregation expression can reflect the changes in the health of the site without displaying individual photovoltaic panels.

[0024] Independent element expression: This level preserves the original outline of the photovoltaic panel and is used for fine-grained visualization at high zoom levels, such as displaying the tilt angle, status, and outline of the panel.

[0025] S3. Write the multi-scale data into a multi-scale data table: Based on the multi-scale optimization results of step S2, the generated various representations are written into the corresponding multi-scale data tables so that they can be called on demand by the vector tiling service at different scaling levels. The three scale representations of irregular features are written into `builds_points`, `buildings_simplified`, and `buildings_detailed`, respectively, while the three scale representations of regular features are written into `solar_panels_points`, `panel_grids`, and `solar_panels`, respectively. After writing, the system checks the data integrity and geometric validity of each table and marks the data preparation as complete so that the containerized tiling service can safely load and generate vector tiles.

[0026] This step ensures that the vector slice generation process can accurately select the data source, improving system reliability.

[0027] S4. Containerized microservices are started in dependency order: This method executes the startup sequence using container orchestration tools like Docker Compose or Kubernetes, in the following order: data preprocessing container, PostGIS database container, Tegola vector tile container, Redis caching container, and front-end rendering container. After successfully completing steps S1 to S3 and writing the preparation flag, the data preprocessing container triggers the startup of the PostGIS container. The Tegola container loads the vector tile configuration file and starts running after detecting the existence of multi-scale tables in the database. The Redis container is ready to cache the tile results.

[0028] This step ensures that the dependencies between modules are correct, avoiding exceptions caused by incorrect service startup order.

[0029] S5. The vector slice container dynamically selects the data table and generates slices based on the scaling level; The Tegola container determines the appropriate data table based on the zoom level included in the client's map request. For irregular features, it uses `buildings_detailed` at zoom levels greater than 16, `buildings_points` at zoom levels less than or equal to 12, and `buildings_simplified` at other zoom levels. For regular feature tables, it uses `solar_panels` at zoom levels greater than 15, `solar_panels_points` at zoom levels less than or equal to 10, and `panel_grids` at other zoom levels. The `!ZOOM!` and `!BBOX!` parameters in the client's map request are replaced by the Tegola container with the current zoom level and tile boundaries. After tile generation, they are stored in a Redis cache as binary vectors.

[0030] This dynamic switching mechanism automatically optimizes the amount of data without changing the client logic, which can significantly improve the smoothness of map zooming.

[0031] S6. Perform precise cache clearing when feature attributes change: When the status of photovoltaic panels, building attributes, or any feature field is updated, a trigger in the database immediately writes the feature's ID and spatial location information to the update event table. After scanning the update event table, the monitoring service performs the following steps: S601. Calculate the tile number corresponding to the feature: The monitoring service calculates the tile range step by step according to all supported zoom levels. First, it obtains the position of the bounding rectangle of the feature, calls the conversion function between latitude and longitude and tile coordinates to obtain a set of coordinates, and then assembles them into a set to be deleted.

[0032] The calculation range for each deletion operation is related to the geometric size of the feature itself, thus maintaining an approximate "fixed-point clearing" effect.

[0033] S602. For rule-based features, additional calculation of the corresponding grid tiles is required: This method uses the function `get_grid_id` to obtain the grid number to which the feature belongs, and then repeats the above tile number calculation process, adding all tiles belonging to the grid to the deletion set. This step ensures that the grid aggregation representation is also updated synchronously.

[0034] S603, Perform cache deletion and automatic rebuild: The listening service sends the slice number to be deleted to the caching service. The caching service deletes the corresponding key-value pairs in Redis one by one. The next time the client accesses these areas, it triggers the path to regenerate the slices to ensure that the latest data is displayed.

[0035] Step S6 ensures real-time synchronization between data updates and visualization results, enabling precise cache control and preventing performance waste caused by global cache failure.

[0036] The method executes steps S1 to S6 sequentially, with each step connected to the next, forming a complete data processing chain from the input of the original elements to the rendering on the client side.

[0037] Example 2 like Figure 2 As shown in the illustration, this embodiment provides an efficient visualization system for vector geographic data based on multi-scale optimization and a containerized architecture. Corresponding to the method described in Embodiment 1, this system is used in smart energy scenarios to classify, represent, publish vector tiles, and manage update-driven precise caches of large-scale geographic features, including photovoltaic panels and buildings. The system is deployed in a containerized runtime environment, with multiple functional modules working collaboratively to form a complete system architecture from raw feature data to client-side rendering.

[0038] The system described in this embodiment mainly includes a data input and intelligent preprocessing module, a containerized microservice deployment module, a client and rendering module, and a collaborative cache management module. In this system, the modules are connected via a virtual network provided by a container orchestration environment. The modules are interconnected through container network interfaces, service name resolution, unified port numbers, and TCP communication. The structure and function of each module are shown below: Data Input and Intelligent Preprocessing Module: The data input and intelligent preprocessing module is deployed in the data preprocessing container and is used to execute all the processing logic of steps S1 to S3 in Example 1, converting the raw geographic feature data into a multi-scale data structure that can be called on demand by the vector tile service. This module includes at least a raw data management unit, a geometric index calculation and classification unit, a multi-scale expression generation unit, and a multi-scale data table writing unit.

[0039] The raw data management unit is used to manage the raw feature data table raw_features, ensuring that each feature contains at least a unique ID (id), a geometry field (geom), a feature category (feature_type), and relevant business attribute fields.

[0040] The geometric index calculation and classification unit calculates the bounding rectangle, fill rate, and number of vertices of the feature in sequence. It uses spatial calculation functions such as ST_Envelope, ST_Area, and ST_NPoints to determine whether the feature belongs to a regular or irregular type, and writes the classification results back to the raw_features table.

[0041] The multi-scale representation generation unit generates six types of representations based on the classification results. Irregular features generate center points (calculated using ST_Centroid), simplified surfaces (simplified using ST_SimplifyPreserveTopology), and detailed surfaces (preserving the original geometry). Regular features generate center points, mesh aggregation representations (generating a 100-meter mesh and statistically analyzing attributes using ST_MakeEnvelope), and independent feature representations (preserving the original photovoltaic panel geometry).

[0042] The multi-scale data table writing unit writes the three types of irregular feature representations into the buildings_points, buildings_simplified, and buildings_detailed tables, respectively, and the three types of regular feature representations into the solar_panels_points, panel_grids, and solar_panels tables, respectively. It also checks the data integrity and geometric validity and writes preparation markers to provide a data foundation for subsequent vector tile generation.

[0043] Through the data input and intelligent preprocessing modules, the system transforms the original feature data into a well-structured multi-scale data system, enabling subsequent vector tile generation and rendering to select the appropriate data source as needed at different scaling levels.

[0044] Containerized microservice deployment module: The containerized microservice deployment module is used to host spatial database services, vector tiling services, caching services, and front-end services. It uses container orchestration tools to achieve automatic deployment and dependency management of each service, corresponding to the system part in step S4 of embodiment 1. This module includes at least a container orchestration unit, a spatial database service unit, a vector tiling service unit, a caching service unit, and a front-end service hosting unit.

[0045] The container orchestration unit is responsible for managing the dependencies of each container. After the data preprocessing container completes S1-S3, PostGIS is started; after the database is loaded, Tegola is started; then Redis and the front-end service are started.

[0046] The spatial database service unit runs on the PostGIS container, stores raw_features and six types of multi-scale data tables, and provides efficient spatial query capabilities.

[0047] The vector tile service unit runs in the Tegola container. It selects the corresponding data table, such as buildings_points, buildings_simplified, buildings_detailed, and solar_panels_points, panel_grids, and solar_panels, based on the scaling level in the client request. It then performs spatial filtering based on the view extent, generates binary vector tiles, and stores them in the cache service.

[0048] The caching service unit runs on Redis and is used to store the generated slices; the front-end service is responsible for providing the slices to the browser for rendering.

[0049] Through the containerized microservice deployment module, the system achieves automatic collaboration and elastic deployment between data preprocessing, database services, vector tiling services, caching services and front-end display services, improving the system's scalability and operational efficiency.

[0050] Client and rendering module: The client and rendering module correspond to the client part in step S5 of embodiment 1, used to load vector tiles on terminals such as browsers and adopt different rendering strategies according to the scaling level. This module includes at least a tile request unit, a vector rendering unit, and a style control unit.

[0051] The tile request unit initiates a tile request to the front-end service unit based on the user's zooming, dragging, and other operations on the map, carrying the current zoom level and the location of the target area.

[0052] The vector rendering unit receives vector tile data from the system and renders elements such as photovoltaic panels and buildings on the client side.

[0053] The style control unit automatically adjusts the symbol size, color, and transparency based on layer type, zoom level, and business attribute information. In the far view, it uses point-level representation; in the medium view, it uses grid aggregation or simplified polygon representation; and in the close view, it uses detailed geometric representation. This allows users to obtain a clear and consistent visualization effect at different view levels.

[0054] Without altering the client's business logic, the client and rendering modules leverage the multi-scale characteristics of vector tile services to optimize data loading, thereby significantly improving the smoothness of map operations.

[0055] Collaborative Cache Management Module: The collaborative cache management module corresponds to step S6 in Implementation Example 1. It is used to achieve precise cache clearing driven by update events when feature attributes change, ensuring the consistency between the map display content and the backend data. This module includes at least a database trigger and event recording unit, a listening service and tile range calculation unit, and a cache update unit.

[0056] When the status of the photovoltaic panel or the building attributes change, the feature ID and geometric information are written into the update event table.

[0057] After reading the update event, the listening service extracts the bounding rectangle of the feature and calculates the corresponding tile number set according to each zoom level. For regular features, it also calculates the tile number corresponding to the grid to which it belongs.

[0058] The slice number to be deleted is passed to Redis, and the corresponding cached data is deleted one by one. When the client accesses the area again, the system regenerates the slice based on the latest data.

[0059] Through the collaborative cache management module, the system only performs targeted clearing of tiles that are actually affected after feature attributes are updated, avoiding global cache invalidation and balancing data real-time performance with system performance.

[0060] In summary, the system described in Embodiment 2, through the division of labor and cooperation among the data input and intelligent preprocessing module, the containerized microservice deployment module, the client and rendering module, and the collaborative cache management module, fully implements the method described in Embodiment 1 at the system architecture level. It achieves multi-scale optimization, containerized deployment, efficient rendering, and precise cache control driven by updates for large-scale vector geographic features. It can provide high-performance, scalable, and easy-to-maintain technical support for the visualization of massive photovoltaic panels and building elements in scenarios such as smart energy.

Claims

1. A method for efficient visualization of vector geographic data based on multi-scale optimization and containerized architecture, characterized in that, Includes the following steps: S1. Data Input and Classification: Calculate the bounding rectangle fill rate and vertex number of the input raw geographic feature data, and classify the features into regular features and irregular features according to preset rules; S2, Intelligent Multi-level Optimization: Generates center point representation, simplified surface representation, and detailed surface representation for irregular elements; generates center point representation, grid aggregation unit representation, and independent element representation for regular elements; S3. Multi-scale data storage: The multi-scale representations obtained in step S2 are stored in multi-scale data tables, including point-level tables, simplified-level tables, grid aggregation tables, and detailed-level tables. S4. Containerized microservice deployment: The data preprocessing container, PostGIS spatial database container and vector tile container are launched sequentially through the container orchestration tool, enabling the tile module to generate vector tiles based on multi-scale optimized geographic feature data; S5. Hierarchical Slice Generation: The vector slice container dynamically selects the corresponding multi-scale data table according to the current scaling level; low scaling level uses center point or aggregate expression, medium scaling level uses simplified surface or mesh aggregate surface, and high scaling level uses detailed and complex geometry. S6. Collaborative Cache Management: When a feature attribute changes, the database trigger sends a message including the feature identifier and spatial information to the listening service. The listening service calculates the range of tile numbers covered by the feature and its grid cell at each scaling level. The cache service performs precise cache deletion based on the tile number.

2. The efficient visualization method for vector geographic data based on multi-scale optimization and containerized architecture according to claim 1, characterized in that, In step S2, the simplified surface representation of irregular elements adopts the topology-preserving Douglas-Puk algorithm.

3. The efficient visualization method for vector geographic data based on multi-scale optimization and containerized architecture according to claim 1, characterized in that, In step S2, the expression of the grid aggregation unit of regular features is generated by dividing the spatial region into regular grids and calculating the number of features within the grid.

4. The efficient visualization method for vector geographic data based on multi-scale optimization and containerized architecture according to claim 1, characterized in that, In step S5, when the zoom level is higher than the preset threshold, a detailed surface or independent feature is loaded; when the zoom level is in the middle range, a simplified surface or mesh aggregation unit is loaded; and when the zoom level is low, a center point representation is loaded.

5. The efficient visualization method for vector geographic data based on multi-scale optimization and containerized architecture according to claim 1, characterized in that, Step S6 involves calculating the slice number based on the bounding rectangle of the updated feature to determine the slice number it covers at each scaling level; for features with grid aggregation, the slice number at the corresponding scaling level is further determined based on the geometric extent of the grid cell to which it belongs.

6. The efficient visualization method for vector geographic data based on multi-scale optimization and containerized architecture according to claim 1, characterized in that, In step S6, the cache service only deletes the cache key-value pair corresponding to the slice number obtained in step S6, without deleting other unaffected caches.

7. A high-efficiency visualization system for vector geographic data based on multi-scale optimization and containerized architecture, used to implement the steps in the methods described in claims 1-6, characterized in that, include: The data input and intelligent preprocessing module is used to receive raw geographic feature data, classify the raw features, and generate differentiated multi-scale representations. The containerized microservice deployment module includes a data preprocessing container, a PostGIS spatial database container, a vector tile container, a Redis cache container, and a front-end service container. These containers are automatically started and work together according to preset dependencies using a container orchestration tool. The client and rendering module are used to load the corresponding vector tiles according to the current view scale and to visualize the different expression levels of regular and irregular elements; The collaborative cache management module includes database triggers, a listener service, and a cache service, which are used to perform precise cache clearing when feature data is updated. This system achieves high-performance collaboration across the entire chain of data optimization, slice generation, cache update, and rendering display through containerized deployment.

8. The efficient visualization system for vector geographic data based on multi-scale optimization and containerized architecture according to claim 7, characterized in that, The data input and intelligent preprocessing module includes an element classification unit, an irregular element multi-scale optimization unit, and a regular element multi-scale optimization unit.

9. A high-efficiency visualization system for vector geographic data based on multi-scale optimization and containerized architecture as described in claim 7, characterized in that, The containerized microservice deployment module defines the dependencies between containers through Docker Compose or Kubernetes, so that the PostGIS container starts after the data preprocessing container completes its optimization tasks, and the vector tile container runs after the PostGIS container starts.

10. A high-efficiency visualization system for vector geographic data based on multi-scale optimization and containerized architecture as described in claim 7, characterized in that, The collaborative cache management module triggers the vector tile container to regenerate tiles for the regions containing the updated features based on the tile number calculated by the listening service.

Images