A method for extracting and automatically synthesizing urban villages based on planar residential area data
By dividing the city into blocks at a 1:5000 scale, encrypting nodes, creating constrained Delaunay triangulations and Thiessen polygon analysis, and combining improved Hausdorf distance, the scientific and efficiency issues of urban village extraction and map generalization were solved, and semantic features were preserved.
Patent Information
- Application Number
- CN202311595480.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-28
- Publication Date
- 2025-12-19
- Estimated Expiration
- 2043-11-28
AI Technical Summary
At a scale of 1:5000, urban villages have unique landforms, with small roof areas, high building density, and narrow roads. Existing technology makes it difficult to draw the houses in the villages separately, resulting in unscientific and unreasonable extraction and map generalization of urban villages.
By acquiring isal residential land data at a scale of 1:5000, dividing the land into blocks using linear road data, densifying nodes, creating constrained Delaunay triangulations, performing Thiessen polygon analysis, calculating area ratio indices, screening out urban village residential areas, and merging and integrating the data by improving Hausdorff distance while preserving semantic features.
It achieves efficient and scientific extraction and automatic integration of urban villages, improves the scientific nature and efficiency of map integration, and maintains the integrity of semantic information.
Smart Images
Figure CN117573789B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of geographic information system spatial analysis and automatic map generalization, and particularly relates to a method for extracting and automatically generalizing urban villages based on area resident data. BACKGROUND
[0002] Urban villages are common in China, and their reconstruction and rebuilding should be scientific and reasonable, which is related to the national economy and people's livelihood and is an urgent problem to be solved in the process of urbanization.
[0003] Basic scale map database is a component of national spatial data infrastructure, which can provide decision support for urban village development, that is, urban villages can be automatically extracted from existing large-scale vector map database, and automatic generalization suitable for the spatial characteristics of urban village resident areas and maintaining their semantic characteristics can be implemented according to the distance relationship between the extracted area resident areas when the map scale is reduced, so as to obtain a small-scale map.
[0004] The maximum scale of the basic scale map database is 1:5000, and at this scale, the urban village resident area has a unique form. Since the urban village has the characteristics of small roof area, high building density, and narrow roads in the area, the cartographer cannot draw each house in the village separately. SUMMARY
[0005] In order to solve the problems in the prior art, the present application provides a method for extracting and automatically generalizing urban villages based on area resident data, which comprises the following steps: acquiring area resident data at a scale of 1:5000, and dividing the area resident data into multiple blocks by using linear road data at a scale of 1:5000; encrypting the nodes on the boundary of each resident area in each block, creating a constrained Delaunay triangulation network based on the encrypted resident area boundary; performing a Voronoi polygon analysis on the constrained Delaunay triangulation network to obtain a Voronoi polygon corresponding to each block; obtaining an area ratio index of each resident area according to the area of each resident area in each block and the area of the Voronoi polygon corresponding to each resident area in each block; and screening the resident areas in each block according to the area ratio index, and taking the resident areas whose area ratio index meets a set threshold range as urban village resident areas. The method is easy to implement, simple and efficient, and the process of extracting and generalizing urban villages is scientific and reasonable.
[0006] The present application adopts the following technical scheme, a method for extracting and automatically generalizing urban villages based on area resident data, comprising:
[0007] Obtaining planar residential area data under a 1:5000 scale, and dividing the planar residential area data into a plurality of blocks by using linear road data corresponding to the 1:5000 scale; each block containing a plurality of residential areas;
[0008] Encrypting nodes on the boundary of each residential area in each block, and creating a constrained Delaunay triangulation based on the encrypted residential area boundary;
[0009] Performing a Voronoi polygon analysis on the residential areas in each block based on the constrained Delaunay triangulation, to obtain a Voronoi polygon corresponding to each block;
[0010] Obtaining an area ratio index of each residential area according to the area of each residential area in each block and the area of the Voronoi polygon corresponding to each residential area in each block;
[0011] Screening the residential areas in each block according to the area ratio index of each residential area, and taking the residential areas with an area ratio index satisfying a set threshold range as village-in-city residential areas.
[0012] Further, after taking the residential areas with an area ratio index satisfying a set threshold range as village-in-city residential areas, the method further comprises:
[0013] Taking the residential areas with an area ratio index not satisfying the set threshold range as non-village-in-city residential areas;
[0014] Calculating an improved Hausdorff distance between two facing line segments of each village-in-city residential area and its adjacent non-village-in-city residential area, and dividing the non-village-in-city residential area into a village-in-city residential area when the improved Hausdorff distance is less than a preset threshold;
[0015] Sequentially traversing all the residential areas in each block to obtain a village-in-city residential area set of each block.
[0016] Further, after obtaining the village-in-city residential area set of each block, the method further comprises:
[0017] Obtaining an improved Hausdorff distance between adjacent village-in-city residential areas or non-village-in-city residential areas in each block, and converting the Hausdorff distance into a map distance under a set scale;
[0018] When the map distance is less than a minimum graphic distance of a map under a set scale, merging the adjacent village-in-city residential areas or non-village-in-city residential areas corresponding to the map distance;
[0019] When the map distance is greater than the minimum graphic distance of the map under the set scale, reselecting the adjacent village-in-city residential areas or non-village-in-city residential areas for comparison;
[0020] The method comprises the following steps: sequentially traversing all residential areas in each block to complete the generalization of the residential areas in each block.
[0021] Further, the improved Hausdorff distance between each village residential area and the adjacent non-village residential area is calculated, specifically as follows:
[0022] Two facing line segments between each village residential area and the adjacent non-village residential area are obtained based on the constrained Delaunay triangulation.
[0023] The Hausdorff distance between each village residential area and the adjacent non-village residential area is calculated according to the two facing line segments between each village residential area and the adjacent non-village residential area.
[0024] Further, the method for obtaining the area ratio index of each residential area is as follows:
[0025] The area ratio index of each residential area is obtained according to the ratio of the area of each residential area to the area of the corresponding Thiessen polygon.
[0026] The method for extracting and automatically generalizing the village from the 1:5000 planar residential area data is easy to implement, simple and efficient, and the extraction of the village and the map generalization process are scientific and reasonable, which solves the problem of obtaining the semantic information of the village from the original residential area data and implementing the map generalization according to the principle of distance priority and semantic retention, and helps to improve the scientificity and efficiency of the 1:5000 residential area map generalization in production. BRIEF DESCRIPTION OF DRAWINGS
[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.
[0028] Figure 1 A village extraction and automatic generalization method based on planar residential area data according to an embodiment of the present application;
[0029] Figure 2 A form comparison schematic diagram of the village and the non-village residential area in a block A in the 1:5000 planar residential area data according to an embodiment of the present application;
[0030] Figure 3A 1:5000 road data main road constitutes a road mesh surface and a 1:5000 residential area data is divided into different residential area groups (blocks) by the road mesh surface schematic diagram of the embodiment of the present application; wherein Figure 3 (a) 1:5000 road data main road constitutes a road mesh surface schematic diagram, Figure 3 (b) 1:5000 residential area group is divided into different residential area groups schematic diagram;
[0031] Figure 4 A residential area group in a block S and all residential area boundary nodes in the block S are locally enlarged by original encryption with a spacing of 3 meters, which is a schematic diagram of the embodiment of the present application; wherein, Figure 4 (a) is a schematic diagram of the original map of the residential area group in the block S, Figure 4 (b) is a local enlarged schematic diagram of a certain residential area node in the block S before encryption; Figure 4 (c) is a local enlarged schematic diagram of a certain residential area node in the block S after encryption;
[0032] Figure 5 A certain block S residential area constraint Delaunay triangle network construction schematic diagram and its local enlarged view are provided as an embodiment of the present application;
[0033] Figure 6 A certain block S residential area is obtained after the Thiessen polygon analysis, and a Thiessen polygon schematic diagram is provided as an embodiment of the present application;
[0034] Figure 7 A schematic diagram of the line segment-oriented acquisition result of two residential areas si and sj based on the constraint Delaunay triangle network is provided as an embodiment of the present application;
[0035] Figure 8 A certain block residential area automatic generalization schematic diagram is provided as an embodiment of the present application. DETAILED DESCRIPTION
[0036] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.
[0037] The flowchart of the village extraction and automatic generalization method based on the planar residential area data is shown in the embodiment of the present application as Figure 1 , which includes:
[0038] acquire the planar residential area data under the scale of 1:5000, and divide the planar residential area data into a plurality of blocks by using the linear road data corresponding to the scale of 1:5000; each block contains a plurality of residential areas;
[0039] Under the scale of 1:5000, the village residential area has a unique form, with small roof area, large building density, narrow roads in the area, and the like, as shown in FIG. 1. Figure 2 At this time, the cartographer cannot draw each house in the village separately, but can only merge a plurality of buildings into a plurality of irregular large polygons which are similar to each other and adjacent to each other. The present application divides the 1:5000 planar residential area data (RESA) into different blocks in different road meshes by using the road mesh surface formed by the main roads in the 1:5000 linear road data (LRDL), as shown in FIG. 2. Figure 3 (a) is a road mesh surface formed by the main roads in the 1:5000 road data, Figure 3 (b) is the RESA data under the scale of 1:5000 superimposed on Figure 3 (a), which is divided into planar residential area groups (i.e., blocks) in different blocks by Figure 3 (a).
[0040] The present application takes a block S in a certain road mesh as an example.
[0041] If the original nodes on the boundary of the residential area are used to establish the constrained Delaunay triangulation network of the residential area, the area of the obtained triangulation network is large, and the shape is rough. The shape of the Thiessen polygon is also rough, and the division of the blank area is rough. Therefore, the accuracy of the area ratio index calculated based on the Thiessen polygon area is low. Therefore, the nodes on the boundary of each residential area in each block must be encrypted, and the constrained Delaunay triangulation network is created based on the encrypted residential area boundary.
[0042] In one embodiment of the present application, the nodes on the boundary of the planar residential area group in S are encrypted to have a spacing of 3 meters, as shown in FIG. 3. Figure 4 (a) is the planar residential area data in the block S; Figure 4 (b) and 4(c) are local enlargements of a certain residential area in S, wherein there are only four nodes (indicated by arrows) on the boundary of the residential area in 4(b), and the nodes on the boundary of the residential area are encrypted in 4(c), and there are a plurality of nodes (a part of which is indicated by arrows).
[0043] The Thiessen polygon analysis is performed on the residential areas in each block based on the constrained Delaunay triangulation network, and the Thiessen polygon corresponding to each block is obtained.
[0044] The application further creates a constrained Delaunay triangulation for the encrypted block S, and performs a Voronoi polygon analysis to obtain a constrained Delaunay triangulation as shown in Figure 5 and a Voronoi polygon T as shown in Figure 6 .
[0045] According to the area of each residential area in each block and the area of the corresponding Voronoi polygon in each block, the area ratio index of each residential area is obtained;
[0046] The application extracts part of the village residential candidate set by calculating the ratio of the planar residential area to the corresponding Voronoi polygon area under the scale of 1:5000. First, the area of all residential areas in the block S and the area of all polygons in the corresponding Voronoi polygon T are obtained, and each si (si∈S) and the corresponding ti (ti∈T) are associated to calculate the ratio of si and the corresponding Voronoi polygon ti area, and the area ratio index AreaRatio is obtained, that is, AreaRatioi=si / ti.
[0047] According to the area ratio index of each residential area, the residential areas in each block are screened, and the residential areas with an area ratio index meeting the set threshold range are regarded as village residential areas
[0048] According to the value of the area ratio index AreaRatio, the residential polygon is divided into two categories, that is, when the area ratio index of the residential area is 1>AreaRatio>=0.519689, it is divided into a village residential candidate set S1, and when the area ratio index of the residential area is AreaRatio<0.519689, it is divided into a non-village residential candidate set S2.
[0049] The embodiment of the application further traverses the village residential areas in S1. During the traversal, the first-order adjacent residential area sj of each village residential area si in S1 is obtained, and the improved Hausdorff distance MHDistance_sisj of si and sj facing the line segment is calculated. If the improved Hausdorff distance MHDistance_sisj of the village residential area and its adjacent residential area is less than 4.4 meters, and sj∈S2 (that is, the residential area sj belongs to a non-village residential area), the corresponding adjacent non-village residential area sj is added to the village residential area S1; otherwise, the residential area sj is not processed, and the adjacent residential area is selected again for judgment.
[0050] As shown in Figure 7The pair of facing line segments of si and sj are shown, and the pair of facing line segments is extracted based on the constrained Delaunay triangulation, and the calculation of the improved Hausdorff distance MHDistance_sisj in the present application is based on the pair of facing line segments, and it can be seen that if the improved Hausdorff distance between the pair of facing line segments is less than 4.4 meters, the residential land originally not belonging to the urban village is included in the urban village candidate set, and the candidate sets S1 and S2 are dynamically updated with the traversal, and it is judged whether the traversal of the set S1 is completed. When the traversal is completed, the obtained S1 is the final selected set of the urban village residential land, and is recorded as SUrbernVillage.
[0051] The improved Hausdorff distance MHDistance_sisj is converted to the distance d on the map according to the scale, for example, d=0.3mm, and the value is compared with the minimum graphic distance d_limit (generally, d_limit=0.2mm), if d<d_limit, the adjacent urban village residential land is merged, otherwise, no change is made. In the integration, the topological relationship between the urban village residential land and the non-urban village residential land should be kept as much as possible to keep the respective semantic characteristics, and if necessary, the distance between them can be appropriately exaggerated to 0.2mm.
[0052] In another embodiment, when the map scale is reduced, for example, from 1:5000 to 1:10000, according to the recorded value of the improved Hausdorff distance MHDistance_sisj between the urban village residential lands, the extracted SUrbernVillage can also be automatically integrated.
[0053] Similarly, the present application also automatically integrates the non-urban village residential land according to the improved Hausdorff distance between the non-urban village residential lands, and the final integrated result is shown as Figure 8 The present application proposes a method for extracting the urban village from the 1:5000 planar residential land data and automatically integrating, which is easy to implement, simple and efficient, and the process of urban village extraction and map integration is scientific and reasonable, which well solves the semantic information of "urban village" from the original residential land data and implements the map integration of "distance priority and semantic maintenance" principle, and is helpful to improve the scientificity and efficiency of the 1:5000 residential land map integration in production.
[0054] The above only describes the preferred embodiments of the present application and should not be used to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims
1. A method for extracting and automatically generalizing urban villages based on planar residential area data, characterized in that, The method comprises the following steps: acquiring planar residential area data under a 1:5000 scale, and dividing the planar residential area data into a plurality of blocks by using linear road data corresponding to the 1:5000 scale; each block contains a plurality of residential areas; encrypting nodes on the boundaries of each residential area in each block, and creating a constrained Delaunay triangulation based on the encrypted residential area boundaries; performing a Voronoi polygon analysis on the residential areas in each block based on the constrained Delaunay triangulation, to obtain a Voronoi polygon corresponding to each residential area in each block; obtaining an area ratio index of each residential area according to the area of each residential area in each block and the area of the Voronoi polygon corresponding to each residential area in each block; screening the residential areas in each block according to the area ratio index of each residential area, and taking the residential areas whose area ratio index meets a set threshold range as village-in-city residential areas; after taking the residential areas whose area ratio index meets the set threshold range as the village-in-city residential areas, the method further comprises the following steps: taking the residential areas whose area ratio index does not meet the set threshold range as non-village-in-city residential areas; calculating an improved Hausdorff distance between each village-in-city residential area and its adjacent non-village-in-city residential area, and dividing the non-village-in-city residential area into a village-in-city residential area when the improved Hausdorff distance is less than a preset threshold; sequentially traversing all the residential areas in each block to obtain a set of village-in-city residential areas in each block; after obtaining the set of village-in-city residential areas in each block, the method further comprises the following steps: obtaining an improved Hausdorff distance between adjacent village-in-city residential areas or non-village-in-city residential areas in each block, and converting the improved Hausdorff distance into a map distance under a set scale; merging the adjacent village-in-city residential areas or non-village-in-city residential areas corresponding to the map distance when the map distance is less than a minimum graphic distance of a map under the set scale; reselecting the adjacent village-in-city residential areas or non-village-in-city residential areas for comparison when the map distance is greater than the minimum graphic distance of the map under the set scale; sequentially traversing all the residential areas in each block to complete the integration of the residential areas in each block; calculating the improved Hausdorff distance between each village-in-city residential area and its adjacent non-village-in-city residential area, specifically comprising the following steps: obtaining two facing line segments between each village-in-city residential area and its adjacent non-village-in-city residential area based on the constrained Delaunay triangulation; calculating the improved Hausdorff distance between each village-in-city residential area and its adjacent non-village-in-city residential area according to the two facing line segments between each village-in-city residential area and its adjacent non-village-in-city residential area. The part of the village-in-city residential land candidate set is extracted by calculating the ratio of the planar residential land to the area of the respective Thiessen polygon at the scale of 1:5000. First, the area of all residential lands in the block S and the area of all polygons in the corresponding Thiessen polygon T are obtained, and the respective si and ti are associated, si∈S, ti∈T. The ratio of si and the corresponding Thiessen polygon ti area is calculated to obtain the area ratio indicator AreaRatio, i.e. AreaRatioi=si / ti; The residential lands in each block are screened according to the area ratio indicator of each residential land, and the residential land whose area ratio indicator meets the set threshold range is taken as the village-in-city residential land; According to the value of the area ratio indicator AreaRatio, the residential land polygon is divided into two categories, i.e. when the area ratio indicator of the residential land is 1>AreaRatio≥0.519689, it is divided into the village-in-city residential land candidate set S1, and when the area ratio indicator of the residential land is AreaRatio<0.519689, it is divided into the non-village-in-city residential land candidate set S2; The village-in-city residential lands in S1 are traversed, and the first-order neighboring residential land sj of each village-in-city residential land si in S1 is obtained during the traversal. The improved Hausdorff distance MHDistance_sisj of the facing line segment of si and sj is calculated. If the improved Hausdorff distance MHDistance_sisj of the village-in-city residential land and its neighboring residential land is less than 4.4 meters, and sj∈S2, the corresponding neighboring non-village-in-city residential land sj is added to the village-in-city residential land S1; otherwise, the residential land sj is not processed, and the neighboring residential land is reselected for judgment. The pair of facing line segments of si and sj is extracted based on the constrained Delaunay triangulation network. The calculation of the improved Hausdorff distance MHDistance_sisj is based on the facing line segment. If the improved Hausdorff distance between the facing line segments is less than 4.4 meters, the residential land originally not belonging to the village-in-city residential land is included in the village-in-city candidate set. The candidate sets S1 and S2 are dynamically updated during the traversal, and whether the traversal of the S1 set is completed is judged. When the traversal is completed, S1 is obtained as the final village-in-city residential land set, which is recorded as SUrbernVillage. The improved Hausdorff distance MHDistance_sisj is converted to the distance d on the map according to the scale, and the value is compared with the minimum graphic distance d_limit. If d<d_limit, the neighboring village-in-city residential lands are merged, otherwise, no change is made, and the topological relationship between the village-in-city residential land and the non-village-in-city residential land is maintained to maintain their respective semantic characteristics.