Method and device for identifying business hotspots based on mercator conversion reclustering

Abstract

The embodiment of the present application relates to the technical field of network optimization, and discloses a service hotspot identification method and device based on Mercator conversion and re-clustering, which comprises the following steps: obtaining the longitude and latitude information of an interest region frame corresponding to a hotspot region according to the attribute of the hotspot region; performing Mercator coordinate conversion on the interest region frame according to the longitude and latitude information, and performing Mercator coordinate completion on the envelope region of the interest region frame to obtain a Mercator coordinate grid set corresponding to the interest region frame; performing clustering on the Mercator coordinate grid set to obtain a clustering result grid set, and performing inverse Mercator conversion on the clustering result grid set to obtain an interest region frame set with an aggregation characteristic; and generating a clustering hotspot region by concave envelope on the interest region frame set. Through the above method, the embodiment of the present application can automatically, efficiently and accurately identify a 5G service hotspot region.

Description

Technical Field

[0001] This invention relates to the field of network optimization technology, specifically to a method and apparatus for identifying service hotspots based on Mercator transformation re-clustering. Background Technology

[0002] As the traffic offloading ratio of 5G networks continues to increase, network optimization is focusing on 5G hotspot areas to ensure a smooth 5G user experience and stimulate user traffic in these areas. To monitor these hotspot areas, existing technologies rely on manual judgment: Option 1 involves subjectively selecting popular local commercial areas as 5G service hotspots; Option 2 involves mapping existing 5G sites onto a GIS layer based on their traffic volume and latitude / longitude, and then manually selecting areas with high 5G traffic concentrations as 5G service hotspots.

[0003] The existing 5G service hotspot identification relies too heavily on manual labor and lacks a high degree of intelligence; the subjective differences in manually selecting hotspot areas are significant, resulting in low accuracy or actual matching; the manual selection and identification process requires the use of GIS layers for searching, which leads to low work efficiency. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention provide a service hotspot identification method and apparatus based on Mercator transformation re-clustering, which overcomes or at least partially solves the above problems.

[0005] According to one aspect of the present invention, a service hotspot identification method based on Mercator transform and re-clustering is provided. The method includes: obtaining latitude and longitude information of the bounding boxes of regions of interest corresponding to the hotspot regions based on hotspot region attributes; performing Mercator coordinate transformation on the bounding boxes of the regions of interest based on the latitude and longitude information, and performing Mercator coordinate completion on the envelope region of the bounding boxes of the regions of interest to obtain a Mercator coordinate grid set corresponding to the bounding boxes of the regions of interest; performing clustering on the Mercator coordinate grid set to obtain a clustering result grid set, and performing an inverse Mercator transform on the clustering result grid set to obtain a set of bounding boxes of regions of interest with clustering characteristics; and performing concave envelope transformation on the set of bounding boxes of regions of interest to generate clustered hotspot regions.

[0006] In one optional approach, obtaining the latitude and longitude information of the boundary of the region of interest corresponding to the hotspot region based on the hotspot region attribute includes: performing polygon crawling based on a first application programming interface (API) provided by an internet map to obtain all information points within the polygon region that satisfy the keywords; performing data cleaning on all the obtained information points to obtain information points that satisfy the conditions of a second application programming interface (API); and traversing and crawling the information points that satisfy the conditions of the second API according to the format of the second API to obtain the full region of interest containing the latitude and longitude information of the boundary.

[0007] In one optional approach, the step of performing Mercator coordinate transformation on the boundary of the region of interest based on the latitude and longitude information includes: calculating the horizontal and vertical coordinates in the Mercator coordinate system based on the latitude and longitude in the Mercator projection; simplifying the grid in the Mercator coordinate system by applying a preset grid transformation formula to obtain a mapping table between the latitude and longitude vertices of the boundary of the region of interest and the Mercator coordinates.

[0008] In one alternative approach, performing an inverse Mercator transformation on the clustering results to obtain a set of interest region bounding boxes with clustering characteristics includes: filtering out a set of interest region bounding boxes that match each grid cell in the clustering result grid set based on the mapping relationship table.

[0009] In one optional approach, performing Mercator coordinate completion on the envelope region of the region of interest to obtain a Mercator coordinate grid set corresponding to the region of interest includes: completing the Mercator grids adjacent to all vertices of the region of interest; and completing the grids in the region of interest that are enveloped by the border grids to obtain the completed Mercator coordinate grid set.

[0010] In one optional approach, the step of clustering the Mercator coordinate graticles to obtain a clustered result graticle set includes: arbitrarily selecting a Mercator graticle from the completed Mercator coordinate graticle set as the current Mercator graticle; calculating all Mercator graticles within a first preset range of latitude and longitude from the center of the current Mercator graticle; if the number of Mercator graticles within the associated region is greater than or equal to the minimum cluster density, then placing all Mercator graticles within the associated region into an associated graticle set, placing the current Mercator graticle into the clustered result graticle set, and assigning class labels; traversing the associated graticle set to extract Mercator graticles and calculating their associated regions; if the number of Mercator graticles within the associated region is greater than or equal to the minimum cluster density... If the number of Mercator rasters is greater than or equal to the minimum cluster density, then the Mercator raster is added to the clustering result raster set and labeled with the same class label as the current Mercator raster; the area of ​​the raster labeled with the same class label in the clustering result raster set is calculated. If the raster area is less than the minimum cluster area, then the raster labeled with the same class label in the clustering result raster set is deleted; otherwise, the raster labeled with the same class label in the clustering result raster set is retained; if the number of Mercator rasters in the association domain is less than the minimum cluster density, then the current loop is exited; all rasters in the completed Mercator raster set are traversed to obtain the clustering result raster set that satisfies the clustering conditions.

[0011] In one optional approach, generating clustered hotspot regions by performing concave envelope generation on the set of interest region bounding boxes includes: finding the vertex with the smallest latitude from the set of interest region bounding boxes; if multiple vertices exist, then finding the vertex with the smallest longitude from among the multiple vertices with the smallest latitude as the starting point; otherwise, using the vertex with the smallest latitude as the starting point, and adding it to the concave envelope vertex set; finding a first vertex and a second vertex using the starting point as the vertex, and drawing circles with the diameters formed by the lines connecting the first vertex and the second vertex to the starting point to obtain a first neighborhood; if there are no other vertices in the first neighborhood, then using the first vertex and the second vertex as concave envelope connection points, and adding them to the concave envelope vertex set; otherwise, drawing circumcircles with the triangles formed by the third vertex, the first vertex, and the second vertex in the first neighborhood, respectively; if there are no other vertices in the circumcircle... The third vertex is then added to the concave envelope vertex set as a concave envelope continuation point. The latest added concave envelope continuation vertex and the previous concave envelope continuation vertex are cyclically selected from the concave envelope vertex set. The first direction angle of the current concave envelope continuation vertex towards the previous concave envelope continuation vertex is obtained, and the second direction angle of the current concave envelope continuation vertex towards the vertices of the region of interest bounding box set is calculated. The fourth vertex corresponding to the largest sector angle formed by the first and second direction angles is found. A second neighborhood is obtained by drawing a circle with the line connecting the fourth vertex and the current concave envelope continuation vertex as the diameter. The latest added concave envelope continuation vertex is obtained based on the second neighborhood until the latest added concave envelope continuation vertex is taken as the starting point. The final concave envelope vertex set is taken as the clustering hotspot region.

[0012] According to another aspect of the present invention, a service hotspot identification device based on Mercator transformation and re-clustering is provided. The device includes: an information acquisition unit, configured to acquire latitude and longitude information of the boundary of the region of interest corresponding to the hotspot region based on the hotspot region attributes; a Mercator transformation unit, configured to perform Mercator coordinate transformation on the boundary of the region of interest based on the latitude and longitude information, and perform Mercator coordinate completion on the envelope region of the boundary of the region of interest to obtain a Mercator coordinate grid set corresponding to the boundary of the region of interest; a Mercator inverse transformation unit, configured to perform clustering on the Mercator coordinate grid set to obtain a clustering result grid set, and perform an inverse Mercator transformation on the clustering result grid set to obtain a set of boundary of the region of interest with clustering characteristics; and a concave envelope generation unit, configured to perform concave envelope generation on the set of boundary of the region of interest to generate clustered hotspot regions.

[0013] According to another aspect of the present invention, a computing device is provided, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0014] The memory is used to store at least one executable instruction, which causes the processor to perform the steps of the above-described service hotspot identification method based on Mercator transformation re-clustering.

[0015] According to another aspect of the present invention, a computer storage medium is provided, the storage medium storing at least one executable instruction, the executable instruction causing the processor to perform the steps of the above-described service hotspot identification method based on Mercator transformation re-clustering.

[0016] This invention, through its embodiments, obtains the latitude and longitude information of the region of interest (ROI) bounding boxes corresponding to the hotspot regions based on their attributes; performs Mercator coordinate transformation on the ROI bounding boxes based on the latitude and longitude information, and completes the envelope region of the ROI bounding boxes with Mercator coordinates to obtain a Mercator coordinate raster set corresponding to the ROI bounding boxes; performs clustering on the Mercator coordinate raster set to obtain a clustering result raster set, and performs an inverse Mercator transformation on the clustering result raster set to obtain a set of ROI bounding box sets exhibiting clustering characteristics; and performs concave envelope generation on the ROI bounding box set to generate clustered hotspot regions, thereby automatically, efficiently, and accurately identifying 5G service hotspot regions.

[0017] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0019] Figure 1 A flowchart illustrating the business hotspot identification method based on Mercator transformation re-clustering provided in an embodiment of the present invention is shown.

[0020] Figure 2 This diagram illustrates the region of interest bounding box of the business hotspot identification method based on Mercator transform re-clustering provided in an embodiment of the present invention.

[0021] Figure 3 This invention provides a schematic diagram of the Mercator projection for a business hotspot identification method based on Mercator transformation and re-clustering, as illustrated in an embodiment of the invention.

[0022] Figure 4This diagram illustrates a simplified Mercator grid for a service hotspot identification method based on Mercator transformation re-clustering provided in an embodiment of the present invention.

[0023] Figure 5 This illustration shows the mapping relationship between the latitude and longitude vertices of the AOI bounding box and the Mercator coordinates of the service hotspot identification method based on Mercator transformation re-clustering provided in an embodiment of the present invention.

[0024] Figure 6 This diagram illustrates the Mercator coordinate transformation of the business hotspot identification method based on Mercator transformation and re-clustering provided in an embodiment of the present invention.

[0025] Figure 7 This diagram illustrates the AOI bounding box vertex completion when the Mercator x-coordinate difference is greater than the y-coordinate difference in the business hotspot identification method based on Mercator transformation and re-clustering provided in this embodiment of the invention.

[0026] Figure 8 This diagram illustrates the AOI border vertex completion when the difference in the Mercator vertical coordinate is greater than the difference in the horizontal coordinate, according to the business hotspot identification method based on Mercator transformation and re-clustering provided in this embodiment of the invention.

[0027] Figure 9 This diagram illustrates a clustering method for identifying business hotspots based on Mercator transformation re-clustering, as provided in an embodiment of the present invention.

[0028] Figure 10 The diagram shows the convex and concave hulls of the service hotspot identification method based on Mercator transformation re-clustering provided in an embodiment of the present invention.

[0029] Figure 11 This diagram illustrates the generation of the concave envelope in the service hotspot identification method based on Mercator transformation re-clustering provided in an embodiment of the present invention.

[0030] Figure 12 This diagram illustrates the hotspot region bounding box formed by the convex hull algorithm of the business hotspot identification method based on Mercator transform re-clustering provided in an embodiment of the present invention.

[0031] Figure 13 This diagram illustrates the hotspot region bounding box formed by the concave envelope algorithm of the business hotspot identification method based on Mercator transformation re-clustering provided in an embodiment of the present invention.

[0032] Figure 14 This diagram illustrates the structure of a service hotspot identification device based on Mercator transformation re-clustering provided in an embodiment of the present invention.

[0033] Figure 15 A schematic diagram of the structure of a computing device provided in an embodiment of the present invention is shown. Detailed Implementation

[0034] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0035] Figure 1 This diagram illustrates a flowchart of a service hotspot identification method based on Mercator transform re-clustering provided by an embodiment of the present invention. This service hotspot identification method based on Mercator transform re-clustering is applied to a server and is primarily used for 5G service hotspot identification. Figure 1 As shown, the business hotspot identification method based on Mercator transform re-clustering includes:

[0036] Step S11: Obtain the latitude and longitude information of the border of the region of interest corresponding to the hot spot region based on the hot spot region attributes.

[0037] In this embodiment of the invention, the user aggregation characteristics of each hotspot area are different. For example, a shopping mall hotspot represents a place for users to relax and have fun, while an industrial park hotspot represents a place for users to work. Therefore, it is necessary to define the attributes of hotspot areas. A hotspot area may have more than one attribute; for example, the attributes of a commercial area hotspot include: office buildings, shopping malls, hotels, etc. An Area of ​​Interest (AOI) represents a set of geographic features with specific attributes. An AOI is formed by connecting a series of latitude and longitude vertices that represent the boundaries of geographic features and is widely used in internet maps. A sub-attribute is a set of multiple Areas of Interest (AOIs) of the same type. An AOI has a key feature: the latitude and longitude information of its boundary represents the outline of the AOI. For example... Figure 2 As shown, by plotting the latitude and longitude information of the AOI border of the Lizheyuan residential community under the residential area attribute of a certain city onto the Geographic Information System (GIS) layer, the location distribution and range of the community can be seen intuitively.

[0038] Different hotspot areas possess different attributes, and a single hotspot area may also have multiple sub-attributes. For example, a commercial district refers to an area where retail businesses are concentrated and transactions are frequent, typically located in the city center, around traffic intersections, bustling streets, and large public facilities. Commercial districts generally include the following four sub-attributes:

[0039] Shopping malls, such as commercial complexes that integrate shopping, leisure, and dining; office buildings, such as XX Ginza and XX Building; hotels, such as XX Hotel and other hotels and guesthouses; supermarkets, such as large well-known supermarkets and local medium and large supermarkets. Commercial areas generally include two attributes: industrial parks and factories.

[0040] Each sub-attribute of a hotspot area has many similar Areas of Interest (AOIs), requiring the acquisition of their specific latitude and longitude bounding box information. The application programming interface (API) and web crawling technology provided by the internet map developer platform can efficiently and in batches acquire the latitude and longitude bounding box information of AOIs. In step S11, optionally, polygon crawling is performed based on the first application programming interface provided by the internet map to acquire all Points of Information (POIs) within the polygon area that satisfy the keywords. Not all acquired POIs satisfy the AOI acquisition conditions; therefore, all acquired POIs are cleaned to acquire those that satisfy the conditions of the second application programming interface. The POIs satisfying the conditions of the second application programming interface are then traversed and crawled according to the format of the second application programming interface to acquire the full range of regions of interest containing their latitude and longitude bounding box information.

[0041] Step S12: Perform Mercator coordinate transformation on the border of the region of interest based on the latitude and longitude information, and perform Mercator coordinate completion on the envelope region of the border of the region of interest to obtain the Mercator coordinate raster set corresponding to the border of the region of interest.

[0042] In this embodiment of the invention, the Mercator projection is a conformal cylindrical projection, which is the most influential and widely used map projection method. In the Mercator projection, meridians are equidistant parallel straight lines, parallels of latitude are also parallel straight lines, and the meridians and parallels of latitude are perpendicular to each other.

[0043] When performing Mercator coordinate transformation on the bounding box of the region of interest based on the latitude and longitude information, the horizontal and vertical coordinates in the Mercator coordinate system are first calculated based on the latitude and longitude information in the Mercator projection. For example... Figure 3 As shown, the Mercator projection is determined by both mkt_x and mkt_y. The following relationship is used to convert latitude and longitude to Mercator coordinates:

[0044]

[0045] Where lon and lat represent longitude and latitude respectively, and mkt_x and mkt_y represent the x-coordinate and y-coordinate in the Mercator coordinate system respectively.

[0046] Then, a preset raster transformation formula is applied to simplify the raster in the Mercator coordinate system, obtaining a mapping table between the latitude and longitude vertices of the region of interest (ROI) border and Mercator coordinates. Based on the previous formula, the latitude and longitude information of the AOI border can be converted to the Mercator coordinate system. The original Mercator coordinate raster (0,0) and the raster (0,1) are only 0.1 meters apart, which is too fine. Therefore, Mercator coordinate raster simplification is necessary, specifically satisfying the following relationship:

[0047]

[0048] Where `mkt_x_old` and `mkt_y_old` represent the x and y coordinates in the Mercator coordinate system before simplification, respectively; `mkt_x_new` and `mkt_y_new` represent the x and y coordinates in the Mercator coordinate system after simplification, respectively; and `grid_level` represents the simplification level. Figure 4 As shown, a 1x1 grid is converted into a 3x3 grid, and a total of 9 grids from (0,0) to (2,2) are converted into the (0,0) grid within the 3x3 grid. Based on the above grid simplification formula, the original grid can be simplified.

[0049] This completes the transformation of the vertices of each AOI bounding box to the Mercator coordinate system. To facilitate subsequent inverse Mercator transformation processing of the clustering results, [further steps are taken]. Figure 5 The format shown maintains a mapping table between the latitude and longitude vertices of the AOI border and Mercator coordinates.

[0050] After converting the latitude and longitude of the AOI bounding box vertices to Mercator coordinates, the region enclosed by the AOI bounding box cannot be converted to Mercator coordinates using formulas because it lacks vertex latitude and longitude. For example... Figure 6 As shown, the three vertices Point1, Point2, and Point3 of the AOI bounding box are transformed to the Mercator coordinate system, but the region enclosed by these vertices is not transformed correctly. Therefore, Mercator coordinate completion is needed for the region enclosed by the bounding box of the region of interest. Specifically, this can be done by first completing the Mercator grids adjacent to all vertices of the bounding box of the region of interest, and then completing the grids within the bounding box of the region of interest that are enclosed by the bounding box grids, resulting in the completed Mercator coordinate grid set.

[0051] When completing the Mercator grid adjacent to all vertices of the region of interest border, the center latitude and longitude of the grid corresponding to each vertex in Mercator coordinates are found sequentially according to the vertex order of the region of interest border. Each vertex grid is connected according to the center latitude and longitude, and the direction angle is calculated. If the direction angle of two adjacent vertices is within the first preset range, the grids bisected by the direction angle are connected sequentially from the starting point as the completed grid. The first preset range is preferably [45°, 135°, 225°, 315°].

[0052] If the orientation angles of two adjacent vertices are not within the first preset range, then determine at least one intersection point between the orientation angles of the two adjacent vertices and the axis with the shorter projection distance between the two adjacent vertices on the Mercator coordinate axis, and the distance of the intersection point from the other axis where the center latitude and longitude of the first of the two adjacent vertices is located. Based on the range of values ​​of the distance, complete the grid corresponding to the intersection point according to the center latitude and longitude of the first vertex.

[0053] Specifically, if the orientation angles of two adjacent vertices are not within [45°, 135°, 225°, 315°], then calculate the difference between the x and y coordinates of the two adjacent vertices in the Mercator coordinate system:

[0054] Diff_x=|mkt_grid2_x-mkt_grid1_x|

[0055] Diff_y=|mkt_grid2_y-mkt_grid1_y|

[0056] If Diff_x > Diff_y, such as Figure 7 As shown, let Angle be the direction angle between two adjacent vertices, and α be the angle between Angle and the Mercator projection y-axis where the starting point is located. i Let x be the Mercator projection of the direction angle Angle and its orientation. i Intersection of axes, Length i Point of intersection i The length of the distance from the center latitude and longitude of mkt_grid1 along the y-axis is Length. i =(0.5+1×i)×tan(α), i∈[1, 2,...,Diff_x-1].

[0057] If 0.5 + 1 × (n - 1) < Length i <0.5+1×n, and Length i-1 If 0.5 + 1 × (n - 1) < Length, then the i-th completed grid cell is the grid cell corresponding to mkt_grid1_x + i and mkt_grid1_y + (n - 1), where n = Diff_x. If 0.5 + 1 × (n - 1) < Length i ≤0.5+1×n, and Length i-1 If ≥0.5+1×(n-1), then the i-th completed grid is the grid corresponding to mkt_grid1_x+i and mkt_grid1_y+n.

[0058] If Diff_x < Diff_y, such as Figure 8 As shown, let Angle be the direction angle between two adjacent vertices, and α be the angle between Angle and the Mercator projection x-axis where the starting point is located. i Let y be the Mercator projection of the direction angle Angle and its orientation. i Intersection of axes, Length i Point of intersection i The length of the distance from the center latitude and longitude of mkt_grid1 on the x-axis is then:

[0059] Length i =(0.5+1×i)×tan(α), i∈[1, 2,…Diff_x-1].

[0060] If 0.5 + 1 × (n - 1) < Length i <0.5+1×n, and Length i-1 If 0.5 + 1 × (n - 1), then the i-th completed grid is the grid corresponding to mkt_grid1_x + (n - 1) and mkt_grid1_y + i; if 0.5 + 1 × (n - 1) < Length i ≤0.5+1×n, and Length i-1 If ≥0.5+1×(n-1), then the i-th completed grid is the grid corresponding to mkt_grid1_x+n and mkt_grid1_y+i.

[0061] This completes the Mercator grid completion for all adjacent vertices, but some grids enclosed by the border grid may still remain incomplete. When completing the grids enclosed by the border grid within the region of interest, the maximum ordinate value Max_y and the minimum ordinate value Min_y are sequentially selected from the border grid set List_bk obtained after completing the Mercator grid completion for all adjacent vertices, and the process is repeated from the minimum ordinate value Min_y to the maximum ordinate value Max_y. Let y... i For the i-th traversal value, select the Mercator projection y-values ​​from the border grid set List_bk that are equal to y. i The Mercator grid is used to count the number of cells. The preferred preset order is counter-clockwise.

[0062] If the quantity is equal to 1, then no Mercator grid needs to be filled in, and the current traversal should be terminated.

[0063] If the number is equal to 2, let mkt_grid2 and mkt_grid1 be the two corresponding Mercator grids, and calculate Diff_x = |mkt_grid2_x - mkt_grid1_x|. If Diff_x > 1, then the Mercator grids between these two Mercator grids are used as the fill grids. If Diff_x = 1, then there are no Mercator grids to fill, and the current iteration ends.

[0064] If the number is greater than 2, let mkt_grid_j be the j-th Mercator grid, and mkt_grid_j+1 be the Mercator grid adjacent to the j-th Mercator grid. Calculate Diff_x = |mkt_grid_j_x - mkt_grid_j+1_x|. If Diff_x > 1, let Angle be the direction angle of mkt_grid_j towards mkt_grid_j+1, and mkt_grid_next and mkt_grid_before be the next and previous adjacent Mercator grids of mkt_grid_j, respectively. next Angle before These are the orientation angles of mkt_grid_j toward the corresponding Mercator grid. If Angle is in Angle next Rotate counterclockwise to Angle before Within the resulting sector, the Mercator grids between mkt_grid_j and mkt_grid_j+1 are used as fill grids; if none exist, no fill is needed. If Diff_x = 1, there are no Mercator grids to fill, and the current iteration ends.

[0065] This completes the completion of all envelope graticles, resulting in the completed Mercator coordinate graticle set corresponding to the bounding box of the region of interest. To achieve automatic identification of hotspot regions, it is necessary to further cluster the transformed subclass AOIs and automatically select the hotspot regions formed by subclass AOIs with clustering characteristics.

[0066] Step S13: Cluster the Mercator coordinate grid set to obtain a clustering result grid set, and perform an inverse Mercator transformation on the clustering result grid set to obtain a set of interest region bounding boxes with clustering characteristics.

[0067] In this embodiment of the invention, optionally, a Mercator grid is randomly selected from the completed Mercator coordinate grid set as the current Mercator grid. All Mercator grids within a first preset range of latitude and longitude from the center of the current Mercator grid are calculated. If the number of Mercator grids within the associated domain is greater than or equal to the minimum clustering density, all Mercator grids within the associated domain are placed into an associated grid set, and the current Mercator grid is placed into a clustering result grid set and labeled with a class. Mercator grids are traversed from the associated grid set to calculate their associated domains. If the number of Mercator grids within the associated domain is greater than or equal to the minimum clustering density... If the cluster density is determined, the Mercator raster is placed into the clustering result raster set and labeled with the same class label as the current Mercator raster. The area of ​​the raster with the same class label in the clustering result raster set is calculated. If the raster area is less than the minimum cluster area, the raster with the same class label in the clustering result raster set is deleted; otherwise, the raster with the same class label in the clustering result raster set is retained. If the number of Mercator rasters in the association domain is less than the minimum cluster density, the current loop is exited. All rasteres in the completed Mercator raster set are traversed to obtain the clustering result raster set that satisfies the clustering conditions. The first preset range can be set as needed.

[0068] like Figure 9 As shown, let Set_mkt_point be the set of Mercator coordinate gratings formed after converting the subclass AOI to Mercator coordinates and completing it; the Corrlt_area is defined as the region whose latitude and longitude distance from the center of the current Mercator raster is within the range R; the clustering atom Key_point is defined as the Mercator raster whose number of Mercator gratings in its Corrlt_area exceeds the minimum clustering density Min_pts. Traverse the Mercator coordinate grating set Set_mkt_point, extract a Mercator raster from the set, and calculate all Mercator gratings within its Corrlt_area.

[0069] If the number of Mercator rasters is greater than or equal to Min_pts, then the current Mercator raster is marked as the clustering atom Key_point, all Mercator rasters within the associated region are added to the associated raster set Set_correlation, the current Mercator raster is added to the clustering result raster set Set_sameClass, and a class label is assigned to the current Mercator raster. Mercator rasters are retrieved from the associated raster set Set_correlation and their associated regions Corrlt_area are calculated. If the number of Mercator rasters is greater than or equal to Min_pts, the raster is added to Set_sameClass and assigned the same class label. Simultaneously, all Mercator rasters within their associated regions are added to the associated raster set Set_correlation. This process is repeated until the number of Mercator rasters in the associated raster set Set_correlation equals 0. Finally, the area S of the raster cells in the clustering result raster set Set_sameClass that are labeled with the same class is calculated. Min_clustering Defined as the minimum cluster area that satisfies the condition of clustering becoming a hotspot region.

[0070] If S≥S Min_clustering If the clustering result raster set Set_sameClass contains Mercator rasters labeled with the same class, then Mercator rasters with the same class label will be retained.

[0071] If S < S Min_clustering If the clustering result raster set Set_sameClass contains Mercator rasters labeled with the same class, then discard them.

[0072] If the number is less than Min_pts, then the Mercator raster non-clustered atom Key_point will exit the loop.

[0073] After completing the traversal of all Mercator rasters in Set_mkt_point, the set of rasters that meet the clustering conditions, Set_sameClass, is taken as the clustering result of hotspot regions.

[0074] Mercator coordinate clustering automatically converges to form hotspot regions. However, since these hotspot regions are composed of Mercator grids, an inverse Mercator transformation is required, converting Mercator grid coordinates into latitude and longitude bounding boxes. Optionally, a set of region of interest (ROI) bounding boxes matching each grid in the clustering result grid set is selected based on the mapping table. That is, the process of converting AOI bounding boxes to Mercator coordinates in step S12 generates a mapping table between the latitude and longitude vertices of the AOI bounding boxes and Mercator coordinates. Based on this mapping table, the set of AOI bounding boxes mapped by the clustering grids is selected by matching it with the clustering result grid set Set-sameClass.

[0075] Step S14: Generate clustered hotspot regions by performing concave occlusion on the set of bounding boxes of the regions of interest.

[0076] In this embodiment of the invention, the final hotspot area is composed of multiple polygonal AOI borders, and a method needs to be found to automatically select the boundaries of the hotspot area. For example... Figure 10 As shown, both convex and concave hulls can be used to select the boundaries of hotspot regions, but compared to convex hulls, concave hulls form regions that better fit reality.

[0077] In this embodiment of the invention, the set of interest region bounding boxes obtained in step S13 is subjected to concave enveloping to generate clustered hotspot regions. Optionally, such as... Figure 11 As shown, the vertex with the smallest latitude is found from the set of bounding boxes of the region of interest, Set_aoi_lonlat. If multiple vertices exist, the vertex with the smallest longitude is found from these multiple vertices with the smallest latitude and used as the starting point. Otherwise, the vertex with the smallest latitude is used as the starting point, Pts. fst And add it to the concave envelope vertex set Set_concave_hull.

[0078] With the aforementioned starting Pts fst Find the first vertex Pts for each vertex. A Second vertex Pts B Taking the first vertex Pts as an example A and the second vertex Pts B The diameter R formed by the line connecting the starting point and the starting point P_fst Draw a circle to obtain the first neighborhood, defined as R neighborhood; if there are no other vertices in the first neighborhood, then set the first vertex Pts. A and the second vertex Pts B As a concave hull continuation point, it is added to the concave hull vertex set Set_concave_hull; otherwise, it is added to the third vertex Pts in the first neighborhood. i The first vertex Pts A and the second vertex Pts B The resulting triangle is circumcircled. If there are no other vertices within the circumcircle, then the third vertex Pts is... i As a concave hull continuation point, it is added to the concave hull vertex set. That is, if there are other vertices in the R neighborhood, then the third vertex Pts appearing in the R neighborhood is selected. i Construct the circumcircle of the triangle formed by the two vertices containing the diameter of the R-neighborhood. If all other vertices of the R-neighborhood are outside the circumcircle, then Pts i As a concave hull continuation point, it is added to the concave hull vertex set Set_concave_hull.

[0079] Iteratively selects the most recently added concave hull continuation vertex Pts from the concave hull vertex set Set_concave_hull. nextThe vertex Pts is connected to the previous concave envelope. before Obtain the current concave envelope continuation vertex Pts next Towards the previous concave envelope, continue to vertex Pts before First direction angle Angle nxt_bf And calculate the current concave envelope continuation vertex Pts respectively. next The second direction angle Angle towards the vertices in the set Set_aoi_lonlat of the region of interest bounding box i Find the first direction angle Angle nxt_bf and the second direction angle Angle i The fourth vertex Pts corresponding to the maximum sector angle is formed. i With the fourth vertex Pts i The vertex Pts connected to the current concave envelope next The second neighborhood, or R-neighborhood, is obtained by drawing a circle with the line connecting the two points as its diameter. Based on the second neighborhood, the latest concave envelope continuation vertex added to the concave envelope vertex set Set_concave_hull is obtained. The specific method is the same as the method for obtaining the latest concave envelope continuation vertex added to the concave envelope vertex set Set_concave_hull from the R-neighborhood, until the latest concave envelope continuation vertex added to the concave envelope vertex set Set_concave_hull is the starting point Pts. fst .

[0080] The final set of concave octagonal vertex sets is used as the clustering hotspot region.

[0081] The Mercator transform-based re-clustering service hotspot identification method of this invention is applied to automatically determine the 5G service hotspots in a certain city, outputting the 5G service hotspots for that city, which are basically consistent with the existing network service hotspots in the city. Figure 12 The bounding boxes for hotspot regions are formed based on the convex hull algorithm. Figure 13 The bounding boxes for hotspot regions formed by the concave envelope algorithm show that the bounding boxes formed by the concave envelope algorithm are more reasonable. Compared with existing technologies, the service hotspot identification method based on Mercator transform re-clustering in this embodiment of the invention can automatically identify 5G service hotspot regions, significantly reducing manual intervention and saving personnel costs; at the same time, the acquisition and processing methods of the process data required for service hotspot identification are encapsulated, greatly improving the identification efficiency and accuracy.

[0082] This invention, through its embodiments, obtains the latitude and longitude information of the region of interest (ROI) bounding boxes corresponding to the hotspot regions based on their attributes; performs Mercator coordinate transformation on the ROI bounding boxes based on the latitude and longitude information, and completes the envelope region of the ROI bounding boxes with Mercator coordinates to obtain a Mercator coordinate raster set corresponding to the ROI bounding boxes; performs clustering on the Mercator coordinate raster set to obtain a clustering result raster set, and performs an inverse Mercator transformation on the clustering result raster set to obtain a set of ROI bounding box sets exhibiting clustering characteristics; and performs concave envelope generation on the ROI bounding box set to generate clustered hotspot regions, thereby automatically, efficiently, and accurately identifying 5G service hotspot regions.

[0083] Figure 14 A schematic diagram of the structure of a service hotspot identification device based on Mercator transform re-clustering according to an embodiment of the present invention is shown. Figure 14 As shown, the service hotspot identification device based on Mercator transformation re-clustering includes: an information acquisition unit 1401, a Mercator transformation unit 1402, a Mercator inverter unit 1403, and a concave envelope generation unit 404. Wherein:

[0084] Information acquisition unit 1401 is used to acquire the latitude and longitude information of the boundary of the region of interest corresponding to the hotspot region based on the hotspot region attributes; Mercator transformation unit 1402 is used to perform Mercator coordinate transformation on the boundary of the region of interest based on the latitude and longitude information, and to perform Mercator coordinate completion on the envelope region of the boundary of the region of interest to obtain a Mercator coordinate grid set corresponding to the boundary of the region of interest; Mercator inverse transformation unit 1403 is used to perform clustering on the Mercator coordinate grid set to obtain a clustering result grid set, and to perform Mercator inverse transformation on the clustering result grid set to obtain a set of boundary of the region of interest with clustering characteristics; concave envelope generation unit 1404 is used to perform concave envelope generation on the set of boundary of the region of interest to cluster hotspot regions.

[0085] In one alternative approach, the information acquisition unit 1401 is used to: perform polygon crawling based on a first application programming interface (API) provided by an internet map to acquire all information points within the polygon area that satisfy the keywords; perform data cleaning on all acquired information points to acquire information points that satisfy the conditions of a second application programming interface (API); and perform traversal crawling on the information points that satisfy the conditions of the second application programming interface according to the format of the second application programming interface to acquire a full region of interest containing border latitude and longitude information.

[0086] In one alternative approach, the Mercator transformation unit 1402 is used to: calculate the abscissa and ordinate in the Mercator coordinate system based on the Mercator projection and the latitude and longitude in the latitude and longitude information respectively; simplify the grid in the Mercator coordinate system by applying a preset grid transformation formula, and obtain a mapping table of latitude and longitude vertices of the boundary of the region of interest and Mercator coordinates.

[0087] In an alternative approach, the Mercator inverter unit 1403 is used to: filter out a set of interest region bounding boxes that match each grid cell in the clustering result grid set based on the mapping table.

[0088] In one alternative embodiment, the Mercator transformation unit 1402 is used to: complete the Mercator grids adjacent to all vertices of the boundary of the region of interest; and complete the grids in the boundary of the region of interest that are enveloped by the boundary grids to obtain the completed Mercator coordinate grid set.

[0089] In one optional manner, the Mercator inverter unit 1403 is configured to: arbitrarily select a Mercator grid from the completed Mercator coordinate grid set as the current Mercator grid; calculate all Mercator grids within a first preset range of latitude and longitude from the center of the current Mercator grid; if the number of Mercator grids within the associated domain is greater than or equal to the minimum clustering density, then add all Mercator grids within the associated domain to an associated grid set, add the current Mercator grid to a clustering result grid set, and assign a class label; traverse the associated grid set to extract Mercator grids and calculate their associated domains; if the number of Mercator grids within the associated domain is greater than... If the cluster density is equal to the minimum cluster density, the Mercator raster is placed into the clustering result raster set and labeled with the same class label as the current Mercator raster. The area of ​​the raster with the same class label in the clustering result raster set is calculated. If the area of ​​the raster is less than the minimum cluster area, the raster with the same class label in the clustering result raster set is deleted; otherwise, the raster with the same class label in the clustering result raster set is retained. If the number of Mercator rasters in the association domain is less than the minimum cluster density, the current loop is exited. All rasters in the completed Mercator raster set are traversed to obtain the clustering result raster set that satisfies the clustering conditions.

[0090] In one optional manner, the concave envelope generation unit 1404 is configured to: find the vertex with the smallest latitude from the set of interest region bounding boxes; if multiple vertices exist, continue to find the vertex with the smallest longitude from the multiple vertices with the smallest latitude as the starting point; otherwise, use the vertex with the smallest latitude as the starting point and add it to the concave envelope vertex set; find the first vertex and the second vertex respectively with the starting point as the vertex, and draw circles with the diameters formed by the lines connecting the first vertex and the second vertex to the starting point to obtain a first neighborhood; if there are no other vertices in the first neighborhood, use the first vertex and the second vertex as concave envelope connection points and add them to the concave envelope vertex set; otherwise, use the triangle formed by the third vertex in the first neighborhood, the first vertex, and the second vertex as the circumcircle; if there are no other vertices in the circumcircle, use the triangle formed by the third vertex in the first neighborhood, the first vertex, and the second vertex as the circumcircle. Points are added to the concave envelope vertex set as concave envelope continuation points. The latest added concave envelope continuation vertex and the previous concave envelope continuation vertex are cyclically selected from the concave envelope vertex set. The first direction angle of the current concave envelope continuation vertex towards the previous concave envelope continuation vertex is obtained, and the second direction angle of the current concave envelope continuation vertex towards the vertices of the region of interest bounding set is calculated respectively. The fourth vertex corresponding to the largest sector angle formed by the first and second direction angles is found. A second neighborhood is obtained by drawing a circle with the line connecting the fourth vertex and the current concave envelope continuation vertex as the diameter. The latest added concave envelope continuation vertex is obtained based on the second neighborhood until the latest added concave envelope continuation vertex is taken as the starting point. The final concave envelope vertex set is taken as the clustering hotspot region.

[0091] This invention, through its embodiments, obtains the latitude and longitude information of the region of interest (ROI) bounding boxes corresponding to the hotspot regions based on their attributes; performs Mercator coordinate transformation on the ROI bounding boxes based on the latitude and longitude information, and completes the envelope region of the ROI bounding boxes with Mercator coordinates to obtain a Mercator coordinate raster set corresponding to the ROI bounding boxes; performs clustering on the Mercator coordinate raster set to obtain a clustering result raster set, and performs an inverse Mercator transformation on the clustering result raster set to obtain a set of ROI bounding box sets exhibiting clustering characteristics; and performs concave envelope generation on the ROI bounding box set to generate clustered hotspot regions, thereby automatically, efficiently, and accurately identifying 5G service hotspot regions.

[0092] This invention provides a non-volatile computer storage medium storing at least one executable instruction that can execute the service hotspot identification method based on Mercator transformation re-clustering in any of the above method embodiments.

[0093] Executable instructions can specifically be used to cause the processor to perform the following operations:

[0094] Based on the hotspot area attributes, obtain the latitude and longitude information of the border of the interest area corresponding to the hotspot area;

[0095] The Mercator coordinate transformation is performed on the border of the region of interest based on the latitude and longitude information, and the Mercator coordinates are completed on the envelope region of the border of the region of interest to obtain the Mercator coordinate raster set corresponding to the border of the region of interest.

[0096] Clustering is performed on the Mercator coordinate grid set to obtain a clustering result grid set, and an inverse Mercator transformation is performed on the clustering result grid set to obtain a set of interest region bounding boxes with clustering characteristics.

[0097] The set of bounding boxes of the regions of interest is used to generate clustered hotspot regions by performing concave occlusion operations.

[0098] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0099] Based on the first application programming interface provided by the Internet map, polygon crawling is performed to obtain all information points within the polygon area that satisfy the keywords.

[0100] All acquired information points are cleaned to obtain information points that meet the conditions of the second application programming interface.

[0101] Information points that meet the conditions of the second application programming interface are crawled according to the format of the second application programming interface to obtain the full region of interest containing border latitude and longitude information.

[0102] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0103] Based on the Mercator projection, the horizontal and vertical coordinates in the Mercator coordinate system are calculated according to the latitude and longitude information.

[0104] The grid in the Mercator coordinate system is simplified by applying a preset grid transformation formula to obtain a mapping table of latitude and longitude vertices of the boundary of the region of interest and Mercator coordinates.

[0105] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0106] Based on the mapping table, a set of interest region borders matching each grid in the clustering result grid set is selected.

[0107] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0108] Complete the Mercator grid adjacent to all vertices of the bounding box of the region of interest;

[0109] The grid cells within the bounding frame of the region of interest are completed to obtain the completed Mercator coordinate grid set.

[0110] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0111] Take any Mercator grid from the Mercator coordinate grid set after completion as the current Mercator grid, and calculate all Mercator grids in the associated domain within the first preset range of latitude and longitude from the center of the current Mercator grid.

[0112] If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then all Mercator rasters within the association domain are placed into the association raster set, and the current Mercator raster is placed into the clustering result raster set and labeled with a class label. Mercator rasters are retrieved from the association raster set and their association domains are calculated. If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then the Mercator raster is placed into the clustering result raster set and labeled with the same class label as the current Mercator raster. The area of ​​rasters labeled with the same class label in the clustering result raster set is calculated. If the area of ​​the rasters is less than the minimum cluster area, then the rasters labeled with the same class label in the clustering result raster set are deleted; otherwise, the rasters labeled with the same class label in the clustering result raster set are retained.

[0113] If the number of Mercator grids in the associated domain is less than the minimum cluster density, then exit the current loop;

[0114] Traverse all rasters in the Mercator raster set formed after completion to obtain the clustering result raster set that meets the clustering conditions.

[0115] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0116] Find the vertex with the smallest latitude from the set of bounding boxes of the region of interest. If there are multiple vertices with the smallest latitude, find the vertex with the smallest longitude from the multiple vertices with the smallest latitude as the starting point. Otherwise, take the vertex with the smallest latitude as the starting point and add it to the set of concave envelope vertices.

[0117] Using the starting point as the vertex, find the first vertex and the second vertex respectively. Draw circles with the diameters formed by the lines connecting the first vertex and the second vertex to the starting point to obtain the first neighborhood. If there are no other vertices in the first neighborhood, then the first vertex and the second vertex are added to the concave envelope vertex set as concave envelope continuation points. Otherwise, use the triangle formed by the third vertex in the first neighborhood, the first vertex, and the second vertex as the circumcircle. If there are no other vertices in the circumcircle, then the third vertex is added to the concave envelope vertex set as concave envelope continuation point.

[0118] The process involves iteratively filtering the newly added concave envelope continuation vertex and the previous concave envelope continuation vertex from the concave envelope vertex set. A first direction angle is obtained from the current concave envelope continuation vertex toward the previous concave envelope continuation vertex. A second direction angle is calculated from the current concave envelope continuation vertex toward the vertices of the region of interest bounding box set. The fourth vertex corresponding to the largest sector angle formed by the first and second direction angles is identified. A second neighborhood is obtained by drawing a circle with the line connecting the fourth vertex and the current concave envelope continuation vertex as its diameter. The newly added concave envelope continuation vertex is then obtained based on this second neighborhood. This process continues until the newly added concave envelope continuation vertex is taken as the starting point.

[0119] The final set of concave octagonal vertex sets is used as the clustering hotspot region.

[0120] This invention, through its embodiments, obtains the latitude and longitude information of the region of interest (ROI) bounding boxes corresponding to the hotspot regions based on their attributes; performs Mercator coordinate transformation on the ROI bounding boxes based on the latitude and longitude information, and completes the envelope region of the ROI bounding boxes with Mercator coordinates to obtain a Mercator coordinate raster set corresponding to the ROI bounding boxes; performs clustering on the Mercator coordinate raster set to obtain a clustering result raster set, and performs an inverse Mercator transformation on the clustering result raster set to obtain a set of ROI bounding box sets exhibiting clustering characteristics; and performs concave envelope generation on the ROI bounding box set to generate clustered hotspot regions, thereby automatically, efficiently, and accurately identifying 5G service hotspot regions.

[0121] A computer program product includes a computer program stored on a computer storage medium, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the service hotspot identification method based on Mercator transformation re-clustering in any of the above method embodiments.

[0122] Executable instructions can specifically be used to cause the processor to perform the following operations:

[0123] Based on the hotspot area attributes, obtain the latitude and longitude information of the border of the interest area corresponding to the hotspot area;

[0124] The Mercator coordinate transformation is performed on the border of the region of interest based on the latitude and longitude information, and the Mercator coordinates are completed on the envelope region of the border of the region of interest to obtain the Mercator coordinate raster set corresponding to the border of the region of interest.

[0125] Clustering is performed on the Mercator coordinate grid set to obtain a clustering result grid set, and an inverse Mercator transformation is performed on the clustering result grid set to obtain a set of interest region bounding boxes with clustering characteristics.

[0126] The set of bounding boxes of the region of interest is used to generate clustered hotspot regions by performing concave occlusion operations.

[0127] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0128] Based on the first application programming interface provided by the Internet map, polygon crawling is performed to obtain all information points within the polygon area that satisfy the keywords.

[0129] All acquired information points are cleaned to obtain information points that meet the conditions of the second application programming interface.

[0130] Information points that meet the conditions of the second application programming interface are crawled according to the format of the second application programming interface to obtain the full region of interest containing border latitude and longitude information.

[0131] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0132] Based on the Mercator projection, the horizontal and vertical coordinates in the Mercator coordinate system are calculated according to the latitude and longitude information.

[0133] The grid in the Mercator coordinate system is simplified by applying a preset grid transformation formula to obtain a mapping table of latitude and longitude vertices of the boundary of the region of interest and Mercator coordinates.

[0134] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0135] Based on the mapping table, a set of interest region borders matching each grid in the clustering result grid set is selected.

[0136] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0137] Complete the Mercator grid adjacent to all vertices of the bounding box of the region of interest;

[0138] The grids within the bounding grid of the region of interest are completed to obtain the completed Mercator coordinate grid set.

[0139] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0140] Take any Mercator grid from the Mercator coordinate grid set after completion as the current Mercator grid, and calculate all Mercator grids in the associated domain within the first preset range of latitude and longitude from the center of the current Mercator grid.

[0141] If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then all Mercator rasters within the association domain are placed into the association raster set, and the current Mercator raster is placed into the clustering result raster set and labeled with a class label. Mercator rasters are retrieved from the association raster set and their association domains are calculated. If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then the Mercator raster is placed into the clustering result raster set and labeled with the same class label as the current Mercator raster. The area of ​​rasters labeled with the same class label in the clustering result raster set is calculated. If the area of ​​the rasters is less than the minimum cluster area, then the rasters labeled with the same class label in the clustering result raster set are deleted; otherwise, the rasters labeled with the same class label in the clustering result raster set are retained.

[0142] If the number of Mercator grids in the associated domain is less than the minimum cluster density, exit the current loop;

[0143] Traverse all rasters in the Mercator raster set formed after completion to obtain the clustering result raster set that meets the clustering conditions.

[0144] In an alternative approach, the executable instructions cause the processor to perform the following operations:

[0145] Find the vertex with the smallest latitude from the set of bounding boxes of the region of interest. If there are multiple vertices with the smallest latitude, find the vertex with the smallest longitude from the multiple vertices with the smallest latitude as the starting point. Otherwise, take the vertex with the smallest latitude as the starting point and add it to the set of concave envelope vertices.

[0146] Using the starting point as the vertex, find the first vertex and the second vertex respectively. Draw circles with the diameters formed by the lines connecting the first vertex and the second vertex to the starting point to obtain the first neighborhood. If there are no other vertices in the first neighborhood, then the first vertex and the second vertex are added to the concave envelope vertex set as concave envelope continuation points. Otherwise, use the triangle formed by the third vertex in the first neighborhood, the first vertex, and the second vertex as the circumcircle. If there are no other vertices in the circumcircle, then the third vertex is added to the concave envelope vertex set as concave envelope continuation point.

[0147] The process involves iteratively filtering the newly added concave envelope continuation vertex and the previous concave envelope continuation vertex from the concave envelope vertex set. A first direction angle is obtained from the current concave envelope continuation vertex toward the previous concave envelope continuation vertex. A second direction angle is calculated from the current concave envelope continuation vertex toward the vertices of the region of interest bounding box set. The fourth vertex corresponding to the largest sector angle formed by the first and second direction angles is identified. A second neighborhood is obtained by drawing a circle with the line connecting the fourth vertex and the current concave envelope continuation vertex as its diameter. The newly added concave envelope continuation vertex is then obtained based on this second neighborhood. This process continues until the newly added concave envelope continuation vertex is taken as the starting point.

[0148] The final set of concave octagonal vertex sets is used as the clustering hotspot region.

[0149] This invention, through its embodiments, obtains the latitude and longitude information of the region of interest (ROI) bounding boxes corresponding to the hotspot regions based on their attributes; performs Mercator coordinate transformation on the ROI bounding boxes based on the latitude and longitude information, and completes the envelope region of the ROI bounding boxes with Mercator coordinates to obtain a Mercator coordinate raster set corresponding to the ROI bounding boxes; performs clustering on the Mercator coordinate raster set to obtain a clustering result raster set, and performs an inverse Mercator transformation on the clustering result raster set to obtain a set of ROI bounding box sets exhibiting clustering characteristics; and performs concave envelope generation on the ROI bounding box set to generate clustered hotspot regions, thereby automatically, efficiently, and accurately identifying 5G service hotspot regions.

[0150] Figure 15 The diagram shows a schematic of the structure of a computing device provided in an embodiment of the present invention. The specific implementation of the device is not limited by the specific embodiments of the present invention.

[0151] like Figure 15 As shown, the computing device may include: a processor 1502, a communications interface 1504, a memory 1506, and a communications bus 1508.

[0152] The processor 1502, communication interface 1504, and memory 1506 communicate with each other via communication bus 1508. Communication interface 1504 is used to communicate with other network elements such as clients or other servers. The processor 1502 executes program 1510, specifically performing the relevant steps in the above-described embodiment of the service hotspot identification method based on Mercator transformation re-clustering.

[0153] Specifically, program 1510 may include program code that includes computer operation instructions.

[0154] Processor 1502 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The device may include one or more processors of the same type, such as one or more CPUs; or different types of processors, such as one or more CPUs and one or more ASICs.

[0155] Memory 1506 is used to store program 1510. Memory 1506 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0156] Specifically, program 1510 can be used to cause processor 1502 to perform the following operations:

[0157] Based on the hotspot area attributes, obtain the latitude and longitude information of the border of the interest area corresponding to the hotspot area;

[0158] The Mercator coordinate transformation is performed on the border of the region of interest based on the latitude and longitude information, and the Mercator coordinates are completed on the envelope region of the border of the region of interest to obtain the Mercator coordinate raster set corresponding to the border of the region of interest.

[0159] Clustering is performed on the Mercator coordinate grid set to obtain a clustering result grid set, and an inverse Mercator transformation is performed on the clustering result grid set to obtain a set of interest region bounding boxes with clustering characteristics.

[0160] The set of bounding boxes of the region of interest is used to generate clustered hotspot regions by performing concave occlusion operations.

[0161] In an alternative manner, the program 1510 causes the processor to perform the following operations:

[0162] Based on the first application programming interface provided by the Internet map, polygon crawling is performed to obtain all information points within the polygon area that satisfy the keywords;

[0163] All acquired information points are cleaned to obtain information points that meet the conditions of the second application programming interface.

[0164] Information points that meet the conditions of the second application programming interface are crawled according to the format of the second application programming interface to obtain the full region of interest containing border latitude and longitude information.

[0165] In an alternative manner, the program 1510 causes the processor to perform the following operations:

[0166] Based on the Mercator projection, the horizontal and vertical coordinates in the Mercator coordinate system are calculated according to the latitude and longitude information.

[0167] The grid in the Mercator coordinate system is simplified by applying a preset grid transformation formula to obtain a mapping table of latitude and longitude vertices of the boundary of the region of interest and Mercator coordinates.

[0168] In an alternative manner, the program 1510 causes the processor to perform the following operations:

[0169] Based on the mapping table, a set of interest region borders matching each grid in the clustering result grid set is selected.

[0170] In an alternative manner, the program 1510 causes the processor to perform the following operations:

[0171] Complete the Mercator grid adjacent to all vertices of the bounding box of the region of interest;

[0172] The grid cells within the bounding frame of the region of interest are completed to obtain the completed Mercator coordinate grid set.

[0173] In an alternative manner, the program 1510 causes the processor to perform the following operations:

[0174] Take any Mercator grid from the Mercator coordinate grid set after completion as the current Mercator grid, and calculate all Mercator grids in the associated domain within the first preset range of latitude and longitude from the center of the current Mercator grid.

[0175] If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then all Mercator rasters within the association domain are placed into the association raster set, and the current Mercator raster is placed into the clustering result raster set and labeled with a class label. Mercator rasters are retrieved from the association raster set and their association domains are calculated. If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then the Mercator raster is placed into the clustering result raster set and labeled with the same class label as the current Mercator raster. The area of ​​rasters labeled with the same class label in the clustering result raster set is calculated. If the area of ​​the rasters is less than the minimum cluster area, then the rasters labeled with the same class label in the clustering result raster set are deleted; otherwise, the rasters labeled with the same class label in the clustering result raster set are retained.

[0176] If the number of Mercator grids in the associated domain is less than the minimum cluster density, then exit the current loop;

[0177] Traverse all rasters in the Mercator raster set formed after completion to obtain the clustering result raster set that meets the clustering conditions.

[0178] In an alternative manner, the program 1510 causes the processor to perform the following operations:

[0179] Find the vertex with the smallest latitude from the set of bounding boxes of the region of interest. If there are multiple vertices with the smallest latitude, find the vertex with the smallest longitude from the multiple vertices with the smallest latitude as the starting point. Otherwise, take the vertex with the smallest latitude as the starting point and add it to the set of concave envelope vertices.

[0180] Using the starting point as the vertex, find the first vertex and the second vertex respectively. Draw circles with the diameters formed by the lines connecting the first vertex and the second vertex to the starting point to obtain the first neighborhood. If there are no other vertices in the first neighborhood, then the first vertex and the second vertex are added to the concave envelope vertex set as concave envelope continuation points. Otherwise, use the triangle formed by the third vertex in the first neighborhood, the first vertex, and the second vertex as the circumcircle. If there are no other vertices in the circumcircle, then the third vertex is added to the concave envelope vertex set as concave envelope continuation point.

[0181] The process involves iteratively filtering the newly added concave envelope continuation vertex and the previous concave envelope continuation vertex from the concave envelope vertex set. A first direction angle is obtained from the current concave envelope continuation vertex toward the previous concave envelope continuation vertex. A second direction angle is calculated from the current concave envelope continuation vertex toward the vertices of the region of interest bounding box set. The fourth vertex corresponding to the largest sector angle formed by the first and second direction angles is identified. A second neighborhood is obtained by drawing a circle with the line connecting the fourth vertex and the current concave envelope continuation vertex as its diameter. The newly added concave envelope continuation vertex is then obtained based on this second neighborhood. This process continues until the newly added concave envelope continuation vertex is taken as the starting point.

[0182] The final set of concave octagonal vertex sets is used as the clustering hotspot region.

[0183] This invention, through its embodiments, obtains the latitude and longitude information of the region of interest (ROI) bounding boxes corresponding to the hotspot regions based on their attributes; performs Mercator coordinate transformation on the ROI bounding boxes based on the latitude and longitude information, and completes the envelope region of the ROI bounding boxes with Mercator coordinates to obtain a Mercator coordinate raster set corresponding to the ROI bounding boxes; performs clustering on the Mercator coordinate raster set to obtain a clustering result raster set, and performs an inverse Mercator transformation on the clustering result raster set to obtain a set of ROI bounding box sets exhibiting clustering characteristics; and performs concave envelope generation on the ROI bounding box set to generate clustered hotspot regions, thereby automatically, efficiently, and accurately identifying 5G service hotspot regions.

[0184] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of the present invention are not directed to any particular programming language. It should be understood that the content of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0185] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0186] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim.

[0187] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0188] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A business hotspot identification method based on Mercator transform re-clustering, characterized in that, The method includes: Based on the hotspot area attributes, obtain the latitude and longitude information of the border of the interest area corresponding to the hotspot area; The Mercator coordinate transformation is performed on the border of the region of interest based on the latitude and longitude information, and the Mercator coordinates are completed on the envelope region of the border of the region of interest to obtain the Mercator coordinate raster set corresponding to the border of the region of interest. Clustering is performed on the Mercator coordinate grid set to obtain a clustering result grid set, and an inverse Mercator transformation is performed on the clustering result grid set to obtain a set of interest region bounding boxes with clustering characteristics. The set of bounding boxes of the regions of interest is used to generate clustered hotspot regions by performing concave occlusion operations.

2. The method according to claim 1, characterized in that, The step of obtaining the latitude and longitude information of the boundary of the region of interest corresponding to the hotspot region based on the hotspot region attributes includes: Based on the first application programming interface provided by the Internet map, polygon crawling is performed to obtain all information points within the polygon area that satisfy the keywords; All acquired information points are cleaned to obtain information points that meet the conditions of the second application programming interface. Information points that meet the conditions of the second application programming interface are crawled according to the format of the second application programming interface to obtain the full region of interest containing border latitude and longitude information.

3. The method according to claim 1, characterized in that, The step of performing Mercator coordinate transformation on the bounding box of the region of interest based on the latitude and longitude information includes: Based on the Mercator projection, the horizontal and vertical coordinates in the Mercator coordinate system are calculated according to the latitude and longitude information. The grid in the Mercator coordinate system is simplified by applying a preset grid transformation formula to obtain a mapping table of latitude and longitude vertices of the boundary of the region of interest and Mercator coordinates.

4. The method according to claim 3, characterized in that, The step of performing an inverse Mercator transform on the clustering results to obtain a set of bounding boxes for regions of interest exhibiting clustering characteristics includes: Based on the mapping table, a set of interest region borders matching each grid in the clustering result grid set is selected.

5. The method according to claim 1, characterized in that, The step of performing Mercator coordinate completion on the envelope region of the region of interest to obtain a Mercator coordinate grid set corresponding to the region of interest includes: Complete the Mercator grid adjacent to all vertices of the bounding box of the region of interest; The grid cells within the bounding frame of the region of interest are completed to obtain the completed Mercator coordinate grid set.

6. The method according to claim 1, characterized in that, The process of clustering the Mercator coordinate grid to obtain a clustering result grid set includes: Take any Mercator grid from the Mercator coordinate grid set after completion as the current Mercator grid, and calculate all Mercator grids in the associated domain within the first preset range of latitude and longitude from the center of the current Mercator grid. If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then all Mercator rasters within the association domain are placed into the association raster set, and the current Mercator raster is placed into the clustering result raster set and labeled with a class label. Mercator rasters are retrieved from the association raster set and their association domains are calculated. If the number of Mercator rasters within the association domain is greater than or equal to the minimum cluster density, then the Mercator raster is placed into the clustering result raster set and labeled with the same class label as the current Mercator raster. The area of ​​rasters labeled with the same class label in the clustering result raster set is calculated. If the area of ​​the rasters is less than the minimum cluster area, then the rasters labeled with the same class label in the clustering result raster set are deleted; otherwise, the rasters labeled with the same class label in the clustering result raster set are retained. If the number of Mercator grids in the associated domain is less than the minimum cluster density, then exit the current loop; Traverse all rasters in the Mercator raster set formed after completion to obtain the clustering result raster set that meets the clustering conditions.

7. The method according to claim 1, characterized in that, The step of generating clustered hotspot regions by performing concave enveloping on the set of interest region bounding boxes includes: Find the vertex with the smallest latitude from the set of bounding boxes of the region of interest. If there are multiple vertices with the smallest latitude, find the vertex with the smallest longitude from the multiple vertices with the smallest latitude as the starting point. Otherwise, take the vertex with the smallest latitude as the starting point and add it to the set of concave envelope vertices. Using the starting point as the vertex, find the first vertex and the second vertex respectively. Draw circles with the diameters formed by the lines connecting the first vertex and the second vertex to the starting point to obtain the first neighborhood. If there are no other vertices in the first neighborhood, then the first vertex and the second vertex are added to the concave envelope vertex set as concave envelope continuation points. Otherwise, use the triangle formed by the third vertex in the first neighborhood, the first vertex, and the second vertex as the circumcircle. If there are no other vertices in the circumcircle, then the third vertex is added to the concave envelope vertex set as concave envelope continuation point. The process involves iteratively filtering the newly added concave envelope continuation vertex and the previous concave envelope continuation vertex from the concave envelope vertex set. A first direction angle is obtained from the current concave envelope continuation vertex toward the previous concave envelope continuation vertex. A second direction angle is calculated from the current concave envelope continuation vertex toward the vertices of the region of interest bounding box set. The fourth vertex corresponding to the largest sector angle formed by the first and second direction angles is identified. A second neighborhood is obtained by drawing a circle with the line connecting the fourth vertex and the current concave envelope continuation vertex as its diameter. The newly added concave envelope continuation vertex is then obtained based on this second neighborhood. This process continues until the newly added concave envelope continuation vertex is taken as the starting point. The final set of concave octagonal vertex sets is used as the clustering hotspot region.

8. A service hotspot identification device based on Mercator transform re-clustering, characterized in that, The device includes: The information acquisition unit is used to acquire the latitude and longitude information of the boundary of the region of interest corresponding to the hot spot area based on the hot spot area attributes; The Mercator transformation unit is used to perform Mercator coordinate transformation on the border of the region of interest based on the latitude and longitude information, and to perform Mercator coordinate completion on the envelope region of the border of the region of interest to obtain a Mercator coordinate grid set corresponding to the border of the region of interest. The Mercator inverter unit is used to cluster the Mercator coordinate grid set to obtain a clustering result grid set, and to perform an inverse Mercator transformation on the clustering result grid set to obtain a set of interest region bounding boxes with clustering characteristics. The concave envelope generation unit is used to generate clustered hotspot regions by performing concave envelope generation on the set of interest region bounding boxes.

9. A computing device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the steps of the service hotspot identification method based on Mercator transformation re-clustering according to any one of claims 1-7.

10. A computer storage medium, characterized in that, The storage medium stores at least one executable instruction that causes the processor to perform the steps of the service hotspot identification method based on Mercator transformation re-clustering according to any one of claims 1-7.

Images