CN115794938A

CN115794938A - Visualization method and device for geographic vector line data and computer equipment

Info

Publication number: CN115794938A
Application number: CN202310065800.0A
Authority: CN
Inventors: 马梦宇; 刘泽邦; 陈荦; 杨岸然; 景宁; 李军; 钟志农; 熊伟; 吴烨
Original assignee: National University of Defense Technology
Current assignee: National University of Defense Technology
Priority date: 2023-02-06
Filing date: 2023-02-06
Publication date: 2023-03-14
Anticipated expiration: 2043-02-06
Also published as: CN115794938B

Abstract

The application relates to a visualization method and device for geographic vector line data and computer equipment. Firstly, a self-adaptive visualization model for drawing multi-level tiles is provided, the model draws the tiles by selecting an optimal visualization method through self-adaptive judgment, and the high-efficiency visualization performance is achieved on each display level, so that multi-level real-time visualization browsing from the whole to the fine of geographic vector line data is realized. Meanwhile, a pixel quad-tree spatial index structure is designed based on a double-layer outer frame structure of geographic vector line elements, and the calculation requirements of judgment selection of a visualization method and tile drawing in a model can be met. Finally, an adaptive visualization algorithm of geographic vector line data is designed based on the constructed pixel quad-tree spatial index structure, and efficient algorithm support can be provided for an adaptive visualization model.

Description

Visualization method and device for geographic vector line data and computer equipment

Technical Field

The present application relates to the field of data visualization technologies, and in particular, to a method and an apparatus for visualizing geographic vector line data, and a computer device.

Background

The geographic vector data model, also called discrete object model, usually describes the shape and position of geographic surface space elements with discrete objects such as points, lines, faces, etc., and describes the range and distribution of geospatial phenomena, and is the core data model of the full-space geographic information system describing spatio-temporal entities. Geographic vector line data is an important component of geographic vector data, and is widely used for describing and expressing various linear features such as roads, administrative divisions, river systems, and building outlines by expressing the real world through a series of sequentially arranged points. With the rapid development of the geographic space data acquisition, crowdsourcing geographic information service and geographic data automatic processing technology of the internet of things, the scale of geographic vector line data shows an explosive growth situation. According to data of a statistical bulletin developed in the transport and transportation industry in 2021, the total mileage of roads in the whole country reaches 528.07 kilometres by the end of 2021, and the quantity scale of related road vectors reaches the million-level scale; by 2022, the scale of geographic vector line elements in OpenStreetMap has reached the hundred million scale. Not only is the data size of the huge and growing geographic vector line data huge, but also most of the data is only presented in a form of table or text, so that a user needs to spend a lot of time and energy on processing and analyzing the data.

The visual analysis technology is an analysis method taking visualization as a means, and expresses a large amount of non-visual, abstract or invisible data generated in scientific calculation in a form of graphic image information intuitively and vividly by means of technologies such as computer graphics, image processing and the like, and provides an effective interactive mode to support a user to explore the data. With the rapid development of network Information technology, a Web Map becomes a currently widely used way for browsing Geographic vector data, meanwhile, a standard Web Tile Map Service (WTMS) is also used for quickly browsing Geographic vector data on the internet, and the Geographic spatial data visualization research based on the Web Map has become a research hotspot in the current Geographic Information Science (GIS) field. By visualizing the geographic vector data on the webpage map, a user can intuitively feel the spatial distribution phenomenon and trend of geographic spatial elements and provide powerful support for exploring and analyzing patterns and rules hidden in the data in the next step.

The visualization of the geographic vector line data is the first step of exploring and mining the geographic vector line data and is also the basis for carrying out the next complex spatial analysis on the geographic vector line data. The classic visual analysis exploration flow needs to be followed in the visualization process: the method can browse the overall distribution of the data at the eagle eye viewing angle and can also zoom individual details of the browsed data. The current mainstream geographic vector data visualization tool and method both adopt a data-oriented computing mode, and the core idea of the tool and method is to perform visualization computing by taking a single vector element as a computing unit. However, with the rapid increase of the data scale, the method can cause the problems of long time consumption for generating the map tile data set, large scale, difficult interactive change of the style and the like, and even a high-performance parallel computing technology is adopted, the multi-level real-time visualization of large-scale geographic vector line data is still difficult to realize.

Visualization of spatial data has become an important means of spatial analysis, and geospatial data visualization helps users to summarize, analyze and reason geospatial data more intuitively. In order to realize the visualization of large-scale geographic vector data, the current mainstream method adopts a solution based on a tile pyramid, wherein the tile pyramid is a multi-resolution and multi-level picture organization model and can support the scaling and browsing of data from rough to fine. However, with the rapid increase of data scale, it is more and more difficult for such grid slicing techniques to meet visualization requirements of seamless data browsing, efficient real-time rendering, and the like, and the specific implementation is as follows: 1) When the data scale is increased, the time consumption for generating the visual tiles is increased, and even the drawing can not be completed under extreme conditions; 2) As the level of the tile pyramid is deepened, the number of tiles required to be generated for browsing high-definition data is exponentially increased, and huge challenges are brought to storage organization; 3) The tile pyramid technology directly converts geographic vector data into static pictures that cannot be changed, and when the geographic vector data changes, a tile data set must be regenerated.

In order to improve visualization efficiency, some single-machine visualization methods achieve visualization quick response through a series of methods for reducing data scale, such as data sampling, aggregation, simplification and the like, for example, a part of representative vector elements or element subsections are selected for visualization, or pre-sampling/aggregation calculation is performed to generate a cached visualization result, and the like. The RS-Tree constructs an R Tree structure for retrieving only the sampling result of the data in the query box; the VisTrees realizes interactive visualization of spatial big data in a histogram form by constructing a multi-dimensional index; tabula pre-calculates to obtain a high-resolution aggregation and sampling result of the data and stores the high-resolution aggregation and sampling result in a database system; the VAS and the POIsam adopt an online sampling technology to acquire a more accurate visualization result of the spatial data; the imMens realizes a data sampling technology based on a database and is applied to interactive visualization of data; the CloudBerry realizes the interactive visualization of the spatial big data by pre-acquiring and caching partial query results. However, none of these operations can obtain a fine visualization effect of a single vector element, and the performance is also limited by the size of the data, and a high-quality visualization effect map cannot be generated.

With the rapid development of the distributed parallel technology, more and more researchers are dedicated to realizing the visual rendering of large-scale geographic vector data by using a distributed big data processing framework Hadoop or Spark. Researchers achieve multi-resolution hotspot distribution graph display of large-scale satellite point data within 5 minutes on the designed space-time index; researchers design a space big data visualization framework HadoopViz based on MapReduce, parallelize a visualization drawing process, and encapsulate five abstract functions to realize various visualization effects such as a scatter diagram, a road network diagram, a thermodynamic diagram and the like. Researchers design a space big data management and visualization framework GeoParkViz based on Spark, and map visualization of hundred million scale space data is realized by adopting five map construction operators. The visualization efficiency is higher than that of HadoopViz, and seamless connection of data management and data visualization can be directly realized in the memory. However, all the above methods need to cache all tile datasets on the display level in advance, so that the visualization resolution of these methods cannot meet the requirement of higher display level (> 10), and cannot support quick change of style. Ghosh et al designs a spatial index AID and AID supporting spatial big data visualization exploration by adopting the idea of generating partial visualization results in advance, and when a user browses, the index directly calls the visualization results cached in advance on an element dense area, and then performs real-time calculation on an element sparse area, thereby realizing high-resolution real-time visualization of spatial big data. However, this method also requires a long time to generate the visualization result in advance, so that it still cannot support interactive change of the data pattern. Finally, these solutions under distributed systems consume a significant amount of computing resources.

Meanwhile, some researches utilize the graphic processing advantage of the GPU to accelerate the geographic vector data visualization processing process, and the visualization performance of such methods is improved compared with the CPU-based visualization method, but the performance requirements for real-time visualization are still difficult to achieve, and the requirements for computing devices are high. The vector tile technology, as a recently popular visualization technology, can support a user to directly render data at a client, flexibly change a display style without generating tiles again, but is limited by tile size and network transmission, and complex drawing synthesis is required when the data size is large.

Disclosure of Invention

Based on this, it is necessary to provide a method, an apparatus and a computer device for visualizing geographic vector line data, so as to meet the requirement of multi-level real-time visualization of large-scale geographic vector line data.

A method of visualizing geovector line data, comprising:

acquiring a geographic vector line data set; the geographic vector line data set comprises a plurality of geographic vector line elements; the geographic vector line element is of a double-layer outer frame structure; the double-layer outer frame structure is a set of a line element minimum outer frame containing geographic vector line elements and a sub-line segment minimum outer frame of each sub-line segment in the geographic vector line elements;

constructing a root node of a pixel quad-tree index structure by taking a global geographic space range as an index range, and performing space topology intersection judgment on a result of recursive quartering of the minimum outer enclosure frame of the line element and the global geographic space range from the root node downwards to generate an index node until the minimum index node containing the minimum outer enclosure frame of the line element is generated;

dividing the index range of the minimum index node to obtain four subspaces, constructing corresponding index nodes when the minimum outer covering frame of the sub-line segment is positioned in the subspaces, and constructing corresponding index nodes when the minimum outer covering frame of the sub-line segment is superposed with the subspaces and the corresponding sub-line segment is superposed with the subspaces until the set highest display level is reached;

when the highest display level is reached, inserting sub-line segments into the R tree linked with the minimum index node corresponding to the intersected subspace to obtain a pixel quadtree index structure;

and acquiring a tile drawing task, searching a target index node in the pixel quad-tree index structure according to the tile drawing task, and selecting a data guide or display guide visualization method according to the node attribute of the target index node to finish drawing the tile map.

An apparatus for visualizing geovector line data, comprising:

the data set acquisition module is used for acquiring a geographic vector line data set; the geographic vector line data set comprises a plurality of geographic vector line elements; the geographic vector line element is of a double-layer outer frame structure; the double-layer outer frame structure is a set of a line element minimum outer frame containing geographic vector line elements and a sub-line segment minimum outer frame of each sub-line segment in the geographic vector line elements;

the root node recursion quartering module is used for constructing a root node of the pixel quadrifork R tree index structure by taking the global geographic space range as an index range, and performing space topology intersection judgment on the results of the line element minimum outer enclosure frame and the global geographic space range recursion quartering downwards from the root node to generate an index node until the minimum index node containing the line element minimum outer enclosure frame is generated;

the minimum index node recursion quartering module is used for dividing the index range of the minimum index node to obtain four subspaces, constructing corresponding index nodes when the minimum outer-wrapping frame of the sub-line segment is positioned in the subspaces, and constructing corresponding index nodes when the minimum outer-wrapping frame of the sub-line segment is superposed with the subspaces and the corresponding sub-line segment is superposed with the subspaces until the set highest display level is reached;

the R tree construction module is used for inserting the sub-line segments into the R tree which is linked with the minimum index node corresponding to the intersected subspace after the highest display level is reached to obtain a pixel quad-tree index structure;

and the self-adaptive visualization module is used for acquiring a tile drawing task, searching a target index node in the pixel quad-tree index structure according to the tile drawing task, and selecting a data-oriented or display-oriented visualization method according to the node attribute of the target index node to finish the drawing of the tile map.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor when executing the computer program implementing the steps of:

when the highest display level is reached, inserting the sub-line segments into the R tree linked with the minimum index node corresponding to the intersected subspace of the sub-line segments to obtain a pixel quad-tree index structure;

According to the method, the device and the computer equipment for visualizing the geographic vector line data, firstly, the two-layer outer-wrapping frame structure of the geographic vector line element is designed, so that when the spatial topology intersection is judged, the minimum outer-wrapping frame of the line element and the minimum outer-wrapping frame of each sub-line segment are sequentially adopted for calculation, after two-layer filtering is realized, whether each sub-line segment is intersected with the spatial topology of the spatial range frame is finally judged, the calculation efficiency is greatly improved, and the construction efficiency of the index structure is greatly improved. Secondly, by designing a pixel quad-tree index structure, unique mapping of a tile/pixel corresponding space range is realized, efficient retrieval of the geographic vector elements in the tile corresponding space range can be supported, the calculation requirements in subsequent multi-level tile drawing can be adapted, and data organization support is provided for a self-adaptive visualization model for multi-level tile drawing. Finally, considering that the visualization method of the display guide is mainly influenced by the display hierarchy, the visualization efficiency on the middle and low display hierarchies is higher, geographic vector elements in the tiles are reduced along with the increase of the display hierarchy, and the efficiency of the visualization method adopting the data guide is also higher. In conclusion, the method and the device can meet the requirement of multi-level real-time visualization of large-scale geographic vector line data.

Drawings

FIG. 1 is a schematic flow chart diagram illustrating a method for visualizing geovector line data in one embodiment;

FIG. 2 is a schematic diagram of a two-level bounding box structure of geographic vector line elements;

FIG. 3 is a diagram of a conventional data-oriented visualization model;

FIG. 4 is a schematic view of a visualization model showing guidance;

FIG. 5 is a schematic diagram of an adaptive visualization model oriented to multi-level tile rendering;

FIG. 6 is a schematic diagram of a pixel quad-tree index structure;

FIG. 7 is a schematic diagram of node information of a pixel quad-tree index structure;

FIG. 8 is a diagram illustrating the operation of the search function SEARCHINODE for the index node;

FIG. 9 is a plot of rendering rate comparison for rendering a tile map using DiDV and DaDV in one embodiment; wherein (a) is in the data set L ₁ The rendering rate in (a) is plotted against the data set L ₂ The rendering rate in (a) is plotted against the rate in (b), and (c) is plotted against the data set L ₃ The rendering rate in (d) is a plot of the data set L ₄ The rendering rate in (e) is a plot of the data set L ₅ The rendering rate in (f) is a plot of the data set L ₆ The drawing rate comparison chart in (1);

FIG. 10 is a graph of tile plots plotted against DiDV versus DaDV plotting rate for another embodiment;wherein (a) is in the data set L ₁ The rendering rate in (a) is plotted against the rate in (b) is plotted against the data set L ₂ The rendering rate in (c) is a plot of the data set L ₃ The drawing rate in (a) is plotted against the drawing rate in (d) is plotted against the data set L ₄ The drawing rate in (e) is a plot of the data set L ₅ The rendering rate in (f) is a plot of the data set L ₆ The drawing rate comparison chart in (1);

FIG. 11 is a graph of tile rendering rate comparison using DiDV and DaDV in another embodiment; wherein (a) is in the data set L ₁ The rendering rate in (a) is plotted against the rate in (b) is plotted against the data set L ₂ The rendering rate in (c) is a plot of the data set L ₃ The drawing rate in (a) is plotted against the drawing rate in (d) is plotted against the data set L ₄ The rendering rate in (e) is a plot of the data set L ₅ The rendering rate in (f) is a plot of the data set L ₆ The drawing rate comparison chart in (1);

FIG. 12 is a total time consumed and tile drawing rate for generating 5000 tiles;

FIG. 13 is a time distribution boxplot for the generation of 5000 tiles;

FIG. 14 is a diagram of an internal structure of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

In one embodiment, as shown in fig. 1, there is provided a method for visualizing geovector line data, comprising the steps of:

step 102, a geographic vector line data set is obtained.

The geographic vector line data set comprises a plurality of geographic vector line elements, the geographic vector line elements are of a double-layer outer frame structure, and the double-layer outer frame structure is a set comprising a line element minimum outer frame of the geographic vector line elements and a sub-line segment minimum outer frame of each sub-line segment in the geographic vector line elements. As shown in fig. 2, a schematic diagram of a two-level bounding box structure of geographic vector line elements is provided.

When creating the index node, the most common way is to use an approximate expression mode to perform space topology intersection determination on the minimum outer-covering frame of the vector line element and a space range frame (i.e. a subspace obtained after four divisions of the index node). However, it is considered that at a low display level, the vector line elements are completely located within the spatial range frame, whereas at a high display level, the vector line elements intersect only a part of the spatial range frame. In order to improve the calculation efficiency, the method traverses the geographic vector line elements in the data set one by one, represents the geographic vector line elements as a double-layer outer-wrapping frame structure shown in fig. 2, sequentially adopts the minimum outer-wrapping frame of the line elements and the minimum outer-wrapping frame of each sub-line segment for calculation when performing subsequent spatial topology intersection judgment, and finally judges whether each sub-line segment is intersected with the spatial range frame in a spatial topology after realizing two-layer filtering.

And 104, constructing a root node of the pixel quad-tree index structure by taking the global geographic space range as an index range, and performing spatial topology intersection judgment on the results of the line element minimum outer covering frame and the global geographic space range recursion quartering downwards from the root node to generate an index node until the minimum index node containing the line element minimum outer covering frame is generated. Different from the index range of the conventional quadtree with the minimum outer package box of the geographic vector data set as the root node, the pixel quadtree takes the global geographic space range as the index range of the root node, and recursion quartering is continuously performed downwards to obtain the index node. The main process is as follows:

1) A root node is created. It is understood that for any geographic vector line data set, the index ranges of the root nodes are global geospatial ranges.

2) And dividing the index range of the root node to obtain four subspaces, and respectively carrying out space topology intersection judgment on the four subspaces and the minimum outsourcing frame of the line element. And when the minimum outsourcing frame of the line element is positioned in a certain subspace, constructing a corresponding index node and continuously carrying out recursive quartering and spatial relationship judgment downwards from the index node until the minimum index node containing the minimum outsourcing frame of the line element is generated.

The index construction process of the method is to judge downwards from a root node, when a double-level structure is not used, each sub-line segment in a vector line element needs to judge downwards from the root node, when the double-level outer frame structure is used, a minimum index node completely containing the line element is firstly generated downwards from the root node, then each sub-line segment of the line element is judged downwards from the minimum index node without judging downwards from the root node, therefore, the judgment times are reduced, and the calculation efficiency is greatly improved.

And 106, dividing the index range of the minimum index node to obtain four subspaces, constructing corresponding index nodes when the minimum outer enclosure of the sub-line segments is positioned in the subspaces, and constructing corresponding index nodes when the minimum outer enclosure of the sub-line segments is superposed with the subspaces and the corresponding sub-line segments are superposed with the subspaces until the set highest display level is reached.

And sequentially selecting the minimum outer covering frame of each sub-line segment, and carrying out space topology intersection judgment on the four subspaces obtained after dividing the minimum outer covering frame and the index range of the minimum index node. If the minimum outsourcing frame of the sub-line segment is positioned in a certain subspace, a corresponding index node is created; if the minimum outer enclosure frame of the sub-line segment is overlapped with certain subspaces, further finely judging whether the sub-line segment is overlapped with the subspaces or not to create corresponding index nodes, and continuing to perform recursive quartering and spatial topological intersection judgment downwards from the newly created index nodes until the set highest display level is reached.

And 108, after the highest display level is reached, inserting the sub-line segments into the R tree linked with the minimum index node corresponding to the intersected subspace to obtain the pixel quadtree index structure. And after all vector line elements are judged, constructing to obtain a final pixel quad-tree index structure.

The method adds the R tree, adjusts the internal index nodes, can support the calculation requirement when adaptively selecting a data-oriented or display-oriented visualization method, and quickly retrieves the corresponding vector line elements through the R tree for quick drawing.

The core idea is that the geographical vector line elements and the results of recursion quartering of a global geographical space range are subjected to space topology intersection judgment (located in or overlapped, namely, containing relation or intersection relation) from the root node from top to bottom to generate index nodes, and after a set highest display level is reached, the geographical vector line elements in the index range of the index nodes are generated into an R tree structure according to the organization mode of an R tree, and finally the pixel quadtree index structure is obtained.

And step 110, acquiring a tile drawing task, searching a target index node in the pixel quad-tree index structure according to the tile drawing task, and selecting a data guide or display guide visualization method according to the node attribute of the target index node to finish drawing the tile map.

In order to browse large-scale geographic vector data on a webpage map, a tile data set on each display level in a tile pyramid needs to be drawn. The current mainstream visualization method takes data as calculation guide, as shown in fig. 3, a visualization model schematic diagram of the traditional data guide is provided, and the core idea is that geographic vector elements within a screen range are obtained through retrieval, then a single vector element is taken as a calculation unit, the vector elements are converted into grid objects convenient for screen display, namely vector rasterization, and finally all the rasterization results are combined to obtain a visualization result. By adopting a data-oriented calculation mode, the visualization performance can be higher when the data size is smaller, however, when the vector element size in the screen range is increased, the calculation amount of visualization processing is also increased sharply, and the multi-level real-time visualization of large-scale geographic vector data is difficult to realize even by adopting a high-performance calculation technology. As shown in fig. 4, a visualization model diagram of display guidance is provided, the model directly uses a screen pixel as a calculation unit, generates a pixel value by determining whether a binary problem of any geographic vector element exists in a spatial range corresponding to the pixel, and finally merges all the pixel values to generate a visualization result. The model converts the geographic vector data drawing calculation problem into a spatial analysis problem with a display pixel as a calculation unit, and further converts the calculation complexity depending on the scale of the vector elements into the calculation complexity depending on the scale of the screen display pixel. When the pixel scale displayed in the viewport is far smaller than the geographic vector data scale on the medium-low display level of the tile pyramid, the model has higher visualization efficiency.

In practical application, when large-scale geographic vector data is zoomed and browsed, visualization tiles on different display levels need to be drawn, and the attribute conditions of each visualization tile are different, for example, the display level of the tile is higher or lower, and the number of geographic vector elements in the tile is more or less. The visualization efficiency of the visualization method using data guidance or display guidance is different. Based on the method, in order to realize real-time visualization of large-scale geographic vector data on each display level, a self-adaptive visualization model facing multi-level tile drawing is provided. The model performs multi-level visual browsing on geographic vector data according to a tile pyramid, provides high-efficiency tile drawing performance to realize real-time visualization of large-scale geographic vector data, combines the advantages of a display guidance and data guidance visualization method, performs self-adaptive judgment on an input tile drawing task, draws a tile map by selecting a visualization algorithm with optimal efficiency, and finally obtains a screen display result, as shown in fig. 5. Meanwhile, in the model, geographic vector data needs to be converted into a specific spatial index structure for supporting the rapid drawing of tiles on multiple levels, and the index structure can provide support for the self-adaptive selection of tile drawing algorithms on one hand and can also meet the calculation requirements of two visualization algorithms of display guidance and data guidance on the other hand. And finally, in the self-adaptive judger, setting corresponding judgment factors, and selecting a visualization algorithm according to the judgment result of the tile drawing task.

In the visualization method of the geographic vector line data, firstly, a self-adaptive visualization model for drawing the multi-level tiles is provided, the model draws the tiles by selecting an optimal visualization method through self-adaptive judgment, and the model has high-efficiency visualization performance on each display level, so that the multi-level real-time visualization browsing from the whole to the fine geographic vector line data is realized. Meanwhile, a pixel element quad-tree spatial index structure is designed based on a double-layer outer frame structure of geographic vector line elements, and the calculation requirements of judgment selection and tile drawing of a visualization method in a model can be met. Finally, an adaptive visualization algorithm of geographic vector line data is designed based on the constructed pixel quad-tree spatial index structure, and efficient algorithm support can be provided for an adaptive visualization model.

In order to draw a tile map on any display level, the data-oriented visualization method needs to quickly search geographic vector line elements which are topologically intersected with the space range corresponding to the tile, and the display-oriented visualization method needs to judge whether the space range corresponding to the pixel element is topologically intersected with any geographic vector line element. The two are similar in that: the spatial topological intersection calculation and judgment need to be performed on the spatial ranges, and the spatial ranges corresponding to the pixels or the spatial ranges corresponding to the tiles used for calculation and judgment all adopt a recursive quartering spatial division criterion (the spatial ranges corresponding to the tiles in the tile pyramid are obtained by continuous recursion quartering of the global spatial range, meanwhile, the specification of each tile is 256 × 256 pixels, and the spatial range corresponding to each pixel can be obtained by recursively dividing the spatial ranges corresponding to the tiles for 8 times). The difference between the two is that: the display-oriented visualization method generates pixel values by directly judging binary problems without retrieving specific geographic vector elements, whereas the data-oriented visualization method requires retrieving specific geographic vector line elements.

Based on the comparative analysis, the invention designs a pixel quad-tree index structure. The pixel quadtree index structure is a quadtree-like structure formed by connecting a plurality of index nodes and is mainly used for dividing the geographic space range. Index nodes are divided into five types: root node, upper left node, upper right node, lower left node, lower right node.

In each Index node, node information such as an Index Range (ISR), geocode | node attribute, and node pointer of the node is recorded. And when the PQR tree reaches the preset highest level, the PQR tree is not divided, and each index node is linked with one R tree on the highest display level of the pixel quad-tree index structure, so that the efficient retrieval of the geographic vector line elements with the spatial topological intersection relation with the index node index range is realized.

The specific description of the pixel quad-tree index structure node information is as follows:

1) The geocode of the index node is obtained by a coding method based on Geohash, which is an address coding technology based on global geographic space range recursive subdivision, can realize unique identification of the geographic space range and is often used for dimension reduction coding representation of the spatial range. As shown in fig. 6, coding is performed downward from the root node, and each downward recursion divides one layer: the coding of the upper left node is '00', the coding of the upper right node is '10', the coding of the lower left node is '01', the coding of the lower right node is '11', the coding lengths of the nodes on different levels are inconsistent, and finally, the Geohash coding in the binary form is obtained and used as the geographic coding of the index node. The encoding method can enable each index node to have a unique number, and the spatial inclusion relationship of the parent-child nodes can be embodied by the inclusion relationship among the geocodes g.

2) For the geographic vector line data, three decision factors of element number, display hierarchy and element length are selected as node attributes.

3) The node pointers mainly comprise four sub-node pointers: an upper left node pointer, an upper right node pointer, a lower left node pointer, and an R-tree pointer. When the node is not on the highest level, the four sub-node pointers respectively point to sub-index nodes corresponding to the four sub-regions after the node index range is divided into four, and the R-tree pointer is null; when the node is at the highest level, the four child node pointers are all null, and the R-tree pointer points to the R-tree connected with the index node.

Fig. 6 shows a schematic diagram of a pixel quad-tree index structure, and fig. 7 is a schematic diagram of node information of the pixel quad-tree index structure.

In the pixel quad-tree index structure, the index range of the index node and the spatial range corresponding to the tile/pixel are uniquely mapped, and the characteristic enables the pixel quad-tree index structure to simultaneously meet the calculation requirements of the self-adaptive judgment, data guidance and display guidance visualization method, and the specific embodiment is as follows: 1) The self-adaptive selection of the visualization method can be completed only by quickly finding out the index nodes corresponding to the tiles and judging the values of the judgment factors recorded in the index nodes; 2) The requirement of data retrieval in the data-oriented visualization method can be met only by quickly finding out index nodes corresponding to the tiles and executing spatial retrieval operation on the R trees connected with the nodes to obtain geographic vector elements in the spatial range corresponding to the tiles; 3) If index nodes are generated in advance during index construction, and correspondence of the spatial range with the geographic spatial elements in the spatial topological intersection relationship is realized, the requirement of binary judgment in the display guide visualization method can be met only by judging whether the index nodes corresponding to the pixels exist after the index is constructed.

After a pixel quad-tree index structure is constructed for a geographic vector line data set, a corresponding index node is generated only by a space range frame which has a space topology intersection relation with geographic vector elements in the data set. Therefore, before the visual drawing of the display guide line or data guide is carried out, for a tile task to be drawn, whether the index node corresponding to the tile exists in the pixel quad-tree index structure or not needs to be judged to determine whether the tile needs to be drawn or not. Meanwhile, when the display-oriented visual rendering is performed, whether the index node corresponding to the pixel exists in the pixel quad-tree index structure or not needs to be judged to determine whether the pixel value needs to be generated or not.

In one embodiment, finding a target index node in a pixel quadtree index structure according to a tile drawing task comprises:

obtaining a first geographical code of a space range frame corresponding to a tile to be drawn according to the tile drawing task, obtaining a pre-constructed index node searching function, and inputting an initial index node and the first geographical code which start to be inquired into the index node searching function;

traversing child node pointers of the initial index nodes one by one in the pixel quad-tree, judging whether child nodes of the initial index nodes exist, directly outputting a blank tile map when the child nodes do not exist, and acquiring second geographic codes of the child nodes when the child nodes exist;

and when the lengths of the second geographic code and the first geographic code are not equal, taking the prefix corresponding digits of the second geographic code and the first geographic code to perform exclusive OR operation, and when the result of the exclusive OR operation is 0, taking the second geographic code and the first geographic code as the input of the index node searching function, recursively executing the index node searching function until the second geographic code and the first geographic code are completely equal and outputting a target index node corresponding to the second geographic code.

As shown in fig. 8, the operation of the search function search of the index node is provided. The invention designs an index node searching function SEARCHINODE, and searches a corresponding index node downwards from a set index node in a 'top-down' manner. The function inputs include the index node from which the query is initiated

Geohash coding of space range frame corresponding to tile/pixel

And when the corresponding index node is inquired, outputting the node, otherwise, outputting the node to be null. The algorithm mainly comprises the following steps:

1) Traversing the child node pointers of the initial node one by one to judge the child nodes

Whether or not it exists;

2) When a child node exists, obtaining

GeoHash coding of

；

3) Judgment of

And

whether the lengths are equal; if the two are not equal, the corresponding digits of the prefixes of the two are taken to carry out XOR operation, which is called prefix matching;

4) If the operation result is 0, the operation will be performed

And

as function input, recursively executing an index node search function;

5) Until when

And

fully equal, output

The index nodes corresponding to the space range frames are output as null in other cases.

In one embodiment, a visualization method for adaptively selecting data guide or display guide according to node attributes of a target index node comprises the following steps:

when the display hierarchy of the target index node is larger than the preset display hierarchy, selecting a data-oriented visualization method; when the display hierarchy of the target index node is not more than the preset display hierarchy and the number of the elements of the target index node is more than the preset element number, selecting a display-oriented visualization method; and when the display hierarchy of the target index node is not more than the preset display hierarchy, the number of the elements of the target index node is not more than the preset element number, and the length of the elements of the target index node is more than the preset element length, selecting a display-oriented visualization method.

When the index node corresponding to the tile exists, the tile needs to be drawn. For the geographic vector line data, a series of decision factors are summarized as follows:

1) And tile display hierarchy: the lower the display hierarchy, the greater the number of line elements, and the longer the visualization process takes.

2) Number of vector line elements within tile: the more line elements, the longer the visualization process takes.

3) Total length of vector line elements within tile: the longer the total length of the line elements, the longer the time consuming visualization process.

At this time, attribute information (including three determination factors, i.e., a display hierarchy, a line element number, and a line element total length) in the index node corresponding to the tile is acquired, and three thresholds for evaluation factors are set.

Because the display-oriented visualization method is mainly influenced by the display hierarchy, the visualization efficiency is higher on the middle and low display hierarchies, geographic vector elements in the tiles are reduced along with the increase of the display hierarchy, and the efficiency of the data-oriented visualization method is higher, the invention uses the display hierarchy to display the hierarchylevelAs the first determination factor, the number of line elements is addedcountTotal length of sum line elementlengthAnd (6) judging. The specific determination flow is as follows:

1) Determining hierarchical informationlevelAnd

the size of (1) whenlevelIs greater than

And if so, selecting a data-oriented visualization method, and otherwise, continuing to judge.

2) Judgment ofcountAnd

the size of (1) whencountIs greater than

And selecting a visualization method for displaying the guide. Otherwise, the judgment is continued.

3) Judgment oflengthAnd with

The size of (1) whenlengthIs greater than

Selecting a display-oriented pixel generation strategy; otherwise, a data-oriented visualization method is selected.

According to the method, a series of threshold values are used for judging and adaptively selecting the visualization method, when the data-oriented visualization method is selected, the display level of the tile and the number scale of the geographic vector elements in the tile are limited within a certain number, the data-oriented visualization method has better visualization performance, and when the tile display level is lower or the number of the geographic vectors in the tile is more, the scheme can adaptively select the display-oriented visualization method, and the pixel value is directly and quickly generated so as to generate the visualization result. In conclusion, the method can have a faster tile drawing speed under different conditions.

In one embodiment, selecting a data-oriented visualization method to complete the rendering of the tile map comprises:

in the pixel quad R tree, performing spatial retrieval operation on the R tree linked with the target index node to obtain target geographic vector line elements in a spatial range frame corresponding to the tile to be drawn, rasterizing the target geographic vector line elements one by one to generate pixel values and outputting a tile map.

Selecting a visualization method of the display guide to complete drawing of the tile map, wherein the visualization method comprises the following steps:

and traversing each pixel of the tile to be drawn, calling an index node searching function to judge whether an index node corresponding to the pixel exists, if so, directly generating a pixel value on the tile map and outputting the final tile map.

In summary, the adaptive selection visualization method can be divided into 5 main steps:

1) And calling an index node searching function, judging whether an index node corresponding to the tile to be drawn exists in the PQR index, if not, directly outputting a blank tile map, otherwise, continuously executing.

2) Node attribute information (level, count, length) of the index node is obtained, judgment is carried out on the node attribute information and a set judgment factor threshold value, and a visualization method is selected in a self-adaptive mode.

3) And if the visualization method of the display guide is selected, traversing each pixel of the tile, continuously calling the index node searching function, judging whether the index node corresponding to the pixel exists, and if so, directly generating a pixel value on the tile map.

4) If a data-oriented visualization method is selected, the spatial range retrieval operation is performed on the R tree linked by the index nodes to obtain geographic vector line elements, and the geographic vector line elements are rasterized one by one to generate pixel values.

5) And finally outputting to obtain the tile map.

The advantages of the adaptive selection visualization method are that: when a display-oriented visualization method is adopted, whether an index node corresponding to a pixel has only a unique judgment path is searched, and a tile map can be quickly generated by adopting simple binary operation; when a data-oriented visualization method is adopted, whether the index nodes corresponding to the tiles have the same unique judgment paths or not is searched, and then vector elements are quickly searched and rasterized to generate a tile map. Eventually, a fast rendering of the tile map at various display levels may be achieved.

It should be understood that, although the steps in the flowchart of fig. 1 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a portion of the steps in fig. 1 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

The following experiments and analyses prove the beneficial effects of the present invention.

In this section, geographic vector line data of different scales are used as experimental subjects, and the experimental data are shown in table 1. L is ₂ As a modemThe collected data is constructed, and the rest of the data sets come from OpenStreetMap (OSM), which is an open source volunteer network map service platform containing geographic vector geographic data. From L ₁ To L ₇ The data size gradually increases, and the spatial scale of the data gradually changes from local to global. L is ₇ Contains data of all line elements of the OSM world, wherein the scale of the line elements reaches billion level. The experimental environment is as follows: the memory is 256 GB, the number of cores is 32, and the processing system is Ubuntu 20.04.

In the experiment, firstly, in order to establish a threshold value of a judgment factor in PAV (Pixel-quad R-tree-based Adaptive Visualization), L is selected in experiment I ₁ -L ₆ As a test Data set, tile rendering performance of a Display-oriented Visualization method (DiDV) and a Data-oriented Visualization method (DaDV) under different values of a determination factor are compared, thereby providing a basis for adaptive determination in PAV. In experiment two, we specifically analyzed the visualization performance of the PAV algorithm on different scale and different spatial scale geographic vector line datasets, where L ₇ The PAV algorithm which is used as a test data set for verifying the generalization capability of the algorithm and finally summarized to obtain the judgment factor threshold value obtained by the experiment can support the multi-level real-time interactive visualization of the large-scale geographic vector data set. In two experiments, because the drawing tasks of each tile are independent, the performance of tile drawing is improved by adopting a mixed parallel mode of MPI and OpenMP, and 8 MPI processes (each process comprises 4 OpenMP threads) are started in all experiments for drawing the tiles. Finally, all the geographic vector line datasets in the experiment were pre-converted into the form of PQR indices to support the PAV algorithm.

TABLE 1 Experimental data set

Experiment one: performance comparison of display-oriented and data-oriented visualization methods

In this section, the threshold values of the decision factors are obtained by mainly comparing the visualization performance of the DiDV and the DaDV, tiles are randomly selected in an experiment, the value of the decision factor of each tile is counted, a tile map is drawn by respectively adopting the DiDV and the DaDV, and for a data set L ₁ -L ₆ We will evaluate the effect of 3 factors of tile display hierarchy, number of line elements, total length of line elements on the performance of the two visualization methods, respectively. In a specific code implementation, diDV writes the code implementation directly, and DaDV is implemented based on the API interface of the mature mapping tool Mapnik that is currently popular.

First, we analyze the performance impact of the display levels on two visualization methods, randomly select 2000 non-blank tiles on each display level through a test program, and calculate the rendering rate of the two methods on each display level, and the experimental result is shown in fig. 9. From the experimental results, it can be seen that: the visualization performance of DiDV is clearly superior to DaDV when the display hierarchy is low, and the performance advantage is more apparent when the data set is large in size, e.g., at L ₄ The rate of drawing of DiDV is 35 times that of DaDV when the display level is 6 on the data set; as display levels increase, the efficiency of DaDV increases dramatically and begins to outperform DiDV, for example at L ₆ The maximum rendering rate on the data set reached 563.8 sheets/sec. In summary, when the display level is low, diDV should be selected, and conversely, daDV should be selected, we use the minimum tile level that completely contains the data set as its spatial scale value, e.g., the spatial scale of the data set in the global space is 0. Displaying a hierarchical threshold when spatial dimensions of the data sets differ

Also different for L ₁ -L ₆ We have observed that

Is 8 to 11 larger than the value of the spatial scale, so we set the value after the spatial scale value of the dataset +11 as in the PAV algorithm

。

Then, we analyze the performance impact of the number of the line elements in the space range corresponding to the tile on the two methods, and we set six different range intervals for the number of the line elements in the tile: 0-400, 400-800, 800-1200, 1200-1600, 1600-2000,>2000. For each range interval, 2000 non-blank tiles were randomly selected, tile rendering rates for DiDV and DaDV were calculated, and the experimental results are shown in fig. 10. From the experimental results, it can be seen that: daDV exhibits better visualization performance than DiDV when the number of line elements is small, however, its tile-rendering rate decreases significantly as the number of line elements increases, while the tile-rendering rate of DiDV maintains a relatively stable level. For the data set L ₁ -L ₆ The experimental results show that the tile drawing rate of DiDV begins to exceed that of DaDV over a range of line element numbers of 400-800, so we use 800 as the line element number threshold in the PAV algorithm

。

Then, we analyze the performance impact of the total length of the inline elements in the space range corresponding to the tile on the two methods, and we set six different range intervals for the total length of the inline elements in the tile: 0-200km, 200-400km, 400-600km, 600-800km,>800km. For each range interval, 2000 non-blank tiles were randomly selected, tile rendering rates for DiDV and DaDV were calculated, and the experimental results are shown in fig. 11. From the experimental results, it can be seen that: daDV exhibits better visualization performance than DiDV when the total line element length is small, however its tile-rendering rate decreases significantly as the total line element length increases, while the tile-rendering rate of DiDV remains at a relatively stable level. For the data set L ₁ -L ₆ The experimental results show that the tile drawing rate of DiDV begins to exceed DaDV over a range of 200-400km in total line element length, so we use 400km as the threshold value of total line element length in PAV algorithm

。

Experiment two: large-scale geographic vector line data multi-level real-time visualization

In this section, it is verified that the PAV algorithm has the capability of supporting multi-level real-time visualization of large-scale geographic vector line data, and in the experiment, we adopt a threshold value determined by experiment one (a)

= 11、

= 800、

= 400 km) into the PAV algorithm, L for all data sets ₁ -L ₇ We randomly displayed 5000 non-empty tiles on the 0-20 display level, and counted the total time spent generating 5000 tiles and the rendering time for generating each visualization tile, respectively. FIG. 12 shows the total time consumed to generate 5000 tiles and the tile rendering rate, and it can be seen from the results that the PAV algorithm maintains relatively stable and efficient visualization performance for all datasets, and the tile rendering rate can reach 366.5 tiles/sec at the lowest, which is much higher than the tile request rate (usually not more than 100 tiles/sec) required to view data in real time. Testing data set L at billion Scale ₇ In the above, the total tile generation time is only 13.6 seconds, and the tile drawing rate reaches 366.5 sheets/second. At the same time, from L ₁ -L ₇ The data set scale is increased remarkably, the total time consumption of tile generation of the algorithm is not increased remarkably, and the tile drawing rate is not in a remarkable descending trend, so that the method has the characteristic of insensitivity to the data set scale and is very suitable for multi-level real-time visualization of large-scale geographic vector line data. FIG. 13 is a time distribution boxed graph of 5000 tiles generated, where-represents the average generation time of the tiles, from which it can be seen that for all datasets, the vast majority of tiles can be rendered within 30 milliseconds, testing dataset L on a billion scale ₇ Above, the maximum generation time of a tile is only 44 milliseconds. Assuming that the current browser sends 100 tile requests, all tasks are drawn through 13 rounds (100 ÷ 8), and the worst case can complete the tasks within 0.57 seconds (13 × 44 ms), which can fully meet the requirement of real-time visualization. In summary, the analysis of the results in FIG. 12 and FIG. 13 shows that PAV is obtainedThe algorithm can efficiently support multi-level real-time visualization of billion-scale geographic vector line data.

In summary, the tile drawing rate can reach nearly hundred tiles per second, and the tile drawing rate is very fast, so that tiles do not need to be generated in advance for storage, only need to be drawn in the memory directly each time, when rendering is required, the tiles are drawn in the memory again, the re-drawing time is very short, and the requirement of interactive response can be met.

Aiming at the hot problem of large-scale geographic vector data visualization in the current GIS field, an efficient multi-level visualization technology is provided for a research object by using large-scale geographic vector line data. The technology adopts a tile pyramid model to display geographic vector line data, takes high-efficiency tile drawing performance as a calculation target, provides a self-adaptive visualization model facing multi-level tile drawing, selects an optimal display guide or data guide visualization method in the model through self-adaptive judgment, and designs a corresponding spatial index structure and a visualization algorithm to support data organization and visualization flow involved in the model. Experiments prove that the technology has efficient tile drawing performance on each display level, can well support multi-level real-time visualization of hundred million-scale geographic vector line data under a single machine condition, and has good application prospect in visual exploration of spatial data. Experiments on billion-scale data sets show that the drawing time of any tile on each display level is no more than 45 milliseconds, the visualization result can be obtained through real-time calculation without caching, multi-level real-time visualization of large-scale geographic vector line data can be well supported under a single-machine condition, and the method has a good application prospect in the field of exploration and analysis of large spatial data.

In one embodiment, there is provided a visualization apparatus of geographic vector line data, including:

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 14. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store photoluminescence spectrum data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of testing a diamond growth apparatus for vacuum leak rate.

It will be appreciated by those skilled in the art that the configuration shown in fig. 14 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In an embodiment, a computer device is provided, comprising a memory storing a computer program and a processor implementing the steps of the method in the above embodiments when the processor executes the computer program.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of visualizing geovector line data, the method comprising:

acquiring a geographic vector line data set; the geographic vector line dataset comprises a plurality of geographic vector line elements; the geographic vector line element is of a double-layer outer frame structure; the double-layer outer covering frame structure is a set of a line element minimum outer covering frame containing geographic vector line elements and a sub-line segment minimum outer covering frame of each sub-line segment in the geographic vector line elements;

constructing a root node of a pixel quad-tree index structure by taking a global geographic space range as an index range, and performing space topology intersection judgment on the results of the line element minimum outer covering frame and the global geographic space range recursion quartering downwards from the root node to generate an index node until a minimum index node containing the line element minimum outer covering frame is generated;

performing recursive quartering on the index range of the minimum index node to obtain four subspaces, constructing a corresponding index node when the minimum outer covering frame of the sub-line segment is positioned in the subspace, and constructing a corresponding index node when the minimum outer covering frame of the sub-line segment is superposed with the subspace and the corresponding sub-line segment is superposed with the subspace until a set highest display level is reached;

after the highest display level is reached, inserting the sub-line segments into the R tree linked with the minimum index node corresponding to the intersected subspace of the sub-line segments to obtain a pixel quad-tree index structure;

2. The method of claim 1, wherein the node information contained in each index node includes a geocode and a node pointer; the node pointers comprise child node pointers;

searching a target index node in the pixel quad-tree index structure according to the tile drawing task, wherein the step of searching the target index node comprises the following steps:

obtaining a first geographic code of a space range frame corresponding to a tile to be drawn according to the tile drawing task;

acquiring a pre-constructed index node search function, and inputting an initial index node and a first geographic code which start to be inquired into the index node search function;

traversing the child node pointers of the initial index node one by one in the pixel quad-tree to judge whether the child nodes of the initial index node exist or not;

when the subnode does not exist, directly outputting a blank tile map, and when the subnode exists, acquiring a second geographic code of the subnode;

when the lengths of the second geographic code and the first geographic code are not equal, taking the prefix corresponding digits of the second geographic code and the first geographic code to perform exclusive OR operation;

and when the result of the exclusive-or operation is 0, taking the second geographical code and the first geographical code as the input of the index node searching function, and recursively executing the index node searching function until the second geographical code and the first geographical code are completely equal and outputting a target index node corresponding to the second geographical code.

3. The method according to claim 2, characterized in that the geocoding of the index node is obtained by a Geohash-based encoding method; the node information contained in each index node also comprises a node type; the node types comprise an upper left node, an upper right node, a lower left node and a lower right node;

the step of obtaining the geocode of the index node through the coding method based on the Geohash comprises the following steps:

coding downwards from a root node, and dividing one layer when recursion downwards is carried out once;

the codes of the upper left node are '00', the code of the upper right node is '10', the code of the lower left node is '01', the code of the lower right node is '11', the code lengths of the index nodes on different display levels are inconsistent, and finally, the binary Geohash code is obtained and used as the geographic code of the index node.

4. The method of claim 3, wherein the node pointers further comprise R-tree pointers; the child node pointers comprise an upper left node pointer, an upper right node pointer, a lower left node pointer and a lower right node pointer;

when the index node is not on the highest display level, the four sub-node pointers respectively point to the sub-index nodes corresponding to the four subspaces obtained after the index range of the index node is divided into four parts, and the R-tree pointer is null;

when the index node is on the highest display level, the four child node pointers are all null, and the R-tree pointer points to the R-tree linked by the index node.

5. The method of claim 1, wherein the node attributes of the index node include display hierarchy, number of elements, and length of elements;

the visualization method for adaptively selecting data guide or display guide according to the node attributes of the target index node comprises the following steps:

when the display level of the target index node is larger than a preset display level, selecting a data-oriented visualization method;

when the display hierarchy of the target index node is not larger than the preset display hierarchy and the number of the elements of the target index node is larger than the preset number of the elements, selecting a display-oriented visualization method;

and selecting a display guide visualization method when the display hierarchy of the target index node is not more than a preset display hierarchy, the number of the elements of the target index node is not more than a preset element number, and the length of the elements of the target index node is more than a preset element length.

6. The method of claim 1, wherein selecting a data-oriented visualization method to complete the rendering of the tile map comprises:

in the pixel quad-tree, performing spatial retrieval operation on the R tree linked with the target index node to obtain a target geographic vector line element in a spatial range frame corresponding to the tile to be drawn;

and rasterizing the target geographic vector line elements one by one to generate pixel values and output a tile map.

7. The method of claim 2, wherein selecting a display-oriented visualization method to complete rendering of the tile map comprises:

and traversing each pixel of the tile to be drawn, calling the index node searching function to judge whether an index node corresponding to the pixel exists, and if so, directly generating a pixel value on the tile map and outputting the final tile map.

8. A method according to claim 2, wherein the pixelized quadtree indexing structure has a highest display level of not less than 8.

9. An apparatus for visualizing geovector line data, the apparatus comprising:

the data set acquisition module is used for acquiring a geographic vector line data set; the geographic vector line dataset comprises a plurality of geographic vector line elements; the geographic vector line element is of a double-layer outer frame structure; the double-layer outer frame structure is a set of a line element minimum outer frame containing geographic vector line elements and a sub-line segment minimum outer frame of each sub-line segment in the geographic vector line elements;

the root node recursion quartering module is used for constructing a root node of a pixel quadrifork R tree index structure by taking a global geographic space range as an index range, and performing spatial topology intersection judgment on the line element minimum outer enclosure frame and a result of the global geographic space range recursion quartering downwards from the root node to generate an index node until the minimum index node containing the line element minimum outer enclosure frame is generated;

a minimum index node recursion quartering module, configured to divide an index range of the minimum index node to obtain four subspaces, construct a corresponding index node when the minimum outer bounding box of the sub-line segment is located in the subspace, and construct a corresponding index node when the minimum outer bounding box of the sub-line segment is overlapped with the subspace and the corresponding sub-line segment is overlapped with the subspace until a set highest display level is reached;

the R tree construction module is used for inserting the sub-line segments into the R tree which is linked with the minimum index node corresponding to the intersected subspace after the highest display level is reached, so as to obtain a pixel quadtree index structure;

10. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program performs the steps of the method according to any of claims 1 to 8.