WO2020073430A1

WO2020073430A1 - Method and system for automatically partitioning urban spatial morphology

Info

Publication number: WO2020073430A1
Application number: PCT/CN2018/116290
Authority: WO
Inventors: 杨俊宴; 曹俊; 刘志成; 王桥; 姚莉; 任刚; 刘志远; 崇志宏
Original assignee: 东南大学
Priority date: 2018-10-12
Filing date: 2018-11-20
Publication date: 2020-04-16
Also published as: CN109492796A

Abstract

A method and system for automatically partitioning urban spatial morphology. The method comprises: acquiring basic data of the urban spatial morphology within a given range, and obtaining a plurality of spatial units (S1); for each spatial unit, performing automatic calculation according to multiple set urban spatial morphological indexes, and generating a summary table of urban spatial morphological features (S2); clustering a matrix corresponding to the summary table of urban spatial morphology features by using an unsupervised clustering algorithm, and in a clustering process, setting different clustering parameters to perform clustering operation for several times, and scoring the clustering results to determine the best clustering result (S3); according to the best clustering result, obtaining the result of automatic partition of urban spatial morphology (S4); and further analyzing the correlation between the indexes and obtaining a key influence index. The method objectively and comprehensively understands the features of the urban spatial morphology, thereby forming the result of the automatic partition and the key influence index, and avoiding the problem such as subjective determination or single dimension of inspection that easily occurs in the traditional method.

Description

Method and system for automatically partitioning urban space shape

Technical field

The invention belongs to the field of urban planning, and relates to a method and system for partitioning urban space morphology, and in particular to a method and system for automatically partitioning urban space morphology by extracting multi-dimensional morphological features of a city's three-dimensional space entity itself.

Background technique

Urban spatial form is the core research object of urban planning discipline. In the process of rapid urbanization, cities continue to "extension" and "grow up", and the urban spatial form has become more and more complicated. Different from administrative divisions, resource divisions, ecological grade divisions, etc., urban space morphology division is to divide blocks of architectural entities in the urban three-dimensional material space. The urban space morphology partition is an important link to analyze the urban space morphology. Through the urban space morphology partition, the inner mechanism of the urban space morphology can be analyzed, and then the structural laws of the urban space can be explored.

At present, the common urban spatial morphological division is determined by single-dimensional morphological indicators such as volume ratio, density, and height, and then the interval threshold is set to partition the indicators. For example, based on the dimension of the maximum building height, the urban spatial form is divided into low-rise zones, multi-storey zones, small high-rise zones, high-rise zones, and super high-rise zones. Based on the dimension of building density, the urban spatial form is divided into low-density zones, medium-density zones, and high-density zones. District etc. This kind of zoning has the problems that it cannot take into account the three-dimensional integrity of the urban spatial form, and the determination of the index interval involves the judgment of the human brain, the arbitrariness of the technical results, and the low work efficiency.

Summary of the invention

Object of the invention: The object of the present invention is to provide a method and system for automatic division of urban space morphology, which can automatically generate urban space morphology division results and further obtain key impact indicators to avoid the failure to take into account the three-dimensional integrity of urban space morphology in traditional operations 3. The determination of the index interval involves the problems of human brain judgment, high randomness of technical results, and low work efficiency.

Technical solution: In order to achieve the above object, an automatic partitioning method of urban space form according to the present invention includes the following steps:

(1) Obtain basic data of urban spatial morphology within a given range, the basic data of urban spatial morphology includes or can extract multiple space units, the outline of the space unit is a closed polygonal space, and the interior includes at least one building Objects, the outline of the building is a closed polygonal space, and the building has layer and / or height information;

(2) For each spatial unit, automatically calculate according to the set of multiple urban spatial morphological indicators to generate a summary table of urban spatial morphological characteristics; where the urban spatial morphological indicators include land area, building density, floor area ratio, maximum base area, maximum At least two of building height, maximum building area, average building height and staggering degree;

(3) The matrix corresponding to the summary table of urban spatial morphological characteristics is clustered using an unsupervised clustering algorithm. During the clustering process, different clustering parameters are set to perform multiple clustering operations, and clustering based on the characteristics of the data is used. Class evaluation indicators score each clustering result to determine the best clustering result;

(4) According to the optimal clustering result, each spatial unit and its corresponding cluster number are obtained as the result of automatic zoning of urban spatial form.

In a preferred embodiment, the method further includes: performing a correlation analysis on the data in the urban space morphology characteristic summary table, and outputting key influence indicators of the automatic division of the urban space morphology.

In a preferred embodiment, the space unit is a vector space unit formed according to block, land use or property rights division. In a preferred embodiment, the space area of the space unit is obtained by geometrically calculating the polygonal space formed by closed polylines; the building density is the sum of the base areas of all buildings in the space unit and the space area of the space unit Ratio; the volume ratio is the ratio of the sum of the product of the base area of all buildings in the space unit and the number of building layers and the area of the space unit; the maximum base area is the base area of all buildings in the space unit Maximum value; The maximum building height is the maximum value of the height of all buildings in the space unit; The maximum building area is the maximum value of the product of the bottom area of all buildings in the space unit and the number of building floors; The average building height is the average of the heights of all buildings in the space unit; the stagger is the variance of the heights of all buildings in the space unit; the base area of the building is a polygonal space formed by closed polylines Obtained by geometric calculation.

In a preferred embodiment, the plurality of urban spatial morphology indicators are the land area of the space unit, building density, floor area ratio, maximum base area, maximum building height, maximum building area, average building height, and stagger.

In a specific embodiment, the unsupervised clustering algorithm is a K-means clustering algorithm based on center points, a hierarchical clustering algorithm based on connectivity, a DBSCAN clustering algorithm based on density, and a GMM-EM clustering based on distribution One or more of the algorithms; the clustering evaluation index is one of Dunn index, Davies-Bouldin index and Silhouette coefficient.

In a preferred embodiment, in the step (3), a K-means clustering algorithm based on Mahalanobis distance is used to cluster the matrix corresponding to the urban space morphological feature summary table. During the clustering process, different clusters are set Number for multiple clustering operations, and use the Silhouette coefficient score to determine the best clustering results; where the Silhouette coefficient calculation method is:

For any point i in the data set, a (i) represents the average distance from data point i to similar data points, b (i) represents the minimum value of the average distance from data point i to other types of data points, s (i) Represents the silhouette coefficient of data point i; averaging the silhouette coefficients of all data points is the overall silhouette coefficient of the clustering result.

In the preferred embodiment, the steps of determining the key impact indicators include: performing correlation calculation on the data in the urban spatial morphological characteristics summary table to obtain a correlation matrix graph; comparing the indexes with the largest sum of absolute values in the correlation matrix graph , As a key impact indicator in the urban spatial form.

In a preferred embodiment, in step (4), after obtaining each spatial unit and its corresponding cluster number, the spatial units of the same cluster number are filled with a color to divide the urban spatial form into The result is presented as a flat geometric image of the colored block.

An automatic zoning system for urban space morphology according to the present invention includes:

The spatial unit acquisition module is used to acquire basic data of the urban spatial form within a given range. The basic data of the urban spatial form includes or is capable of extracting a plurality of spatial units. The outline of the spatial unit is a closed polygonal space. At least one building is included, the outline of the building is a closed polygonal space, and the building is provided with layer and / or height information;

The spatial morphology index calculation module is used to automatically calculate according to the set multiple urban spatial morphology indexes for each spatial unit to generate a summary table of urban spatial morphology characteristics; among them, the urban spatial morphology indexes include land area, building density, floor area ratio, At least two of the maximum base area, maximum building height, maximum building area, average building height, and staggered degree;

The cluster evaluation module is used to cluster the matrix corresponding to the urban space morphological characteristics summary table by unsupervised clustering algorithm. During the clustering process, different clustering parameters are set for multiple clustering operations, and the data-based The cluster evaluation index of its own characteristics scores each cluster result to determine the best cluster result;

And, the result output module is used to obtain each spatial unit and its corresponding cluster number according to the optimal clustering result as the result of the automatic division of the urban spatial form; or output the key impact indicators of the automatic division of the urban spatial form The key impact indicators of the urban spatial morphology partition are output; the key impact indicators are obtained by performing correlation analysis on the data in the urban spatial morphological characteristics summary table.

An automatic zoning system for urban space morphology according to the present invention includes at least one computer device. The computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. The computer program When being loaded into the processor, the automatic division method of the urban space form is realized.

Beneficial effect: A method and system for automatic division of urban space morphology proposed by the present invention takes urban space morphology as a three-dimensional whole, comprehensively considers multi-dimensional morphological features, and uses computer algorithms to incorporate multiple urban space morphological features for clustering operations. It can achieve the maximum approximation of the "whole" measurement of the spatial form, rather than the "side" measurement; the output of the partition result is also based on the characteristic parameters of the data such as the Silhouette coefficient, automatically converging the result to the state where the clustering result distribution is optimal, Avoid arbitrary judgment caused by human brain judgment. Furthermore, the key impact indicators in the urban spatial morphology zoning are further determined, which is conducive to grasping the characteristics of key spatial morphological indicators in the planning and design of the research scope, and then grasping the urban spatial morphological characteristics of the research scope as a whole. The invention avoids the problems that the three-dimensional integrity of the urban spatial form cannot be taken into account in the traditional operation, the lack of scientific basis for the determination of key impact indicators, the determination of the indicator interval involves the judgment of the human brain, the technical results are arbitrary, and the working efficiency is low. In order to be able to analyze the inner mechanism of urban space form, and then explore the structural laws of urban space, it provides a reliable rational scientific basis.

BRIEF DESCRIPTION

FIG. 1 is a flowchart of a method according to an embodiment of the present invention.

2 is a schematic diagram of the result of performing the K-means clustering algorithm on the matrix corresponding to the Dengfeng urban spatial morphological feature summary table in an embodiment of the present invention, where (a) k = 2, (b) k = 4, (c ) k = 6, (d) k = 10.

FIG. 3 is a graph showing the results of scoring clustering results using Silhouette coefficients in an embodiment of the present invention.

4 is a flowchart of a method according to another embodiment of the present invention.

FIG. 5 is a correlation matrix diagram using Pearson Correlation Matrix operation in an embodiment of the present invention.

detailed description

The technical solution of the present invention will be further described in detail below with reference to the drawings and embodiments.

As shown in FIG. 1, an embodiment of the present invention discloses an automatic urban space morphing method, which mainly includes the following steps:

S1: Obtain the basic data of urban spatial morphology within a given range, and obtain a number of spatial units as the basis of the partition algorithm from the basic data.

In this step, the basic database of urban spatial form can be obtained through government departments or other data providing platforms. The basic database includes several spatial units according to block, land or property rights; map or image data can also be processed through geographic information processing software. Processing, combined with the basic data of urban buildings to process and obtain. The space unit is usually a block (a geometric shape surrounded by urban roads), and it can also be land (further division of blocks according to the "Urban Land Classification and Planning and Construction Land Standards") or property rights (spatial range formed based on land ownership). The outline is a polygon space formed by closed polylines. The data format can be dwg format or shp format. There are generally buildings inside the space unit, and the outline of the building is also a polygon space composed of closed polylines. The data format can also be dwg format or shp format. Each building inside the space unit is marked with layer information (such as 1, 2,3, ...) and / or altitude information. When there is no height information, the height information (layer number information * 3m) may be calculated from the number of layers.

S2: For each spatial unit, it is automatically calculated according to the set multiple urban spatial morphological indicators, and a summary table of urban spatial morphological characteristics is generated.

In this step, the urban spatial morphological indicators are useful indicators of the characteristics of the space unit, such as land area, building density, floor area ratio, maximum base area, maximum building height, maximum building area, average building height, and stagger. The geometric area of the outline of the space unit can be obtained by geometric calculation of the polygon space formed by the closed polyline, and the land area of the space unit can be expressed by the geometric area of the outline; similarly, the geometric space of the polygon space formed by the closed polyline can also be obtained The geometric area of the building outline serves as the base area of each building in the space unit.

Building density is the ratio of the sum of the bottom areas of all buildings in the space unit to the area of the space unit; the volume ratio is the sum of the product of the base areas of all the buildings in the space unit and the number of building floors and the area of the space unit Ratio; the maximum base area is the maximum value of the bottom area of all buildings in the space unit; the maximum building height is the maximum value of the height of all buildings in the space unit; the maximum building area is the bottom of all buildings in the space unit The maximum value of the product of the area and the number of building layers; the average building height is the average of the heights of all buildings in the space unit; the stagger is the variance of the heights of all buildings in the space unit.

Generally, at least two of the above indexes are selected to calculate the spatial morphological characteristics summary table of each spatial unit. In the actual operation process, the user can choose a combination or give a default index combination, such as: building density, floor area ratio, for the city Comprehensive measurement of the intensity category in the form of space; the combination of the maximum building height, average building height, and staggered degree to comprehensively measure the height category in the urban spatial form; the maximum base area, maximum building height, and maximum building area combination, for the city The architectural iconic categories in the spatial form are comprehensively measured; all index combinations are used to "approximate" the overall conceptual category of the object of the urban spatial form to the greatest extent. It should be noted that the urban spatial form is a complex and comprehensive concept or object. Each dimension of the index can be regarded as a description of the object from one side; as the number of indicators increases, the label describing the object of the urban spatial form The more there are, the more dimensions are defined and described; the land area, building density, floor area ratio, maximum base area, maximum building height, maximum building area, average building height, staggering degree listed in the embodiments of the present invention The most common and common indicators for describing the urban spatial form of a spatial unit; but the present invention is not limited to the above indicators and does not exclude other possible indicators. For example, the mutual calculation between the indicators listed in the present invention can infinitely define new index of.

S3: Cluster the matrix corresponding to the urban spatial morphological characteristics summary table, perform multiple clustering operations by setting different clustering parameters, and use the clustering evaluation index based on the characteristics of the data to select the best clustering result.

In this step, common unsupervised clustering algorithms can be used for clustering operations, such as K-means clustering based on center points, hierarchical clustering based on connectivity, DBSCAN clustering based on density, and GMM-EM based on distribution Clustering. In the clustering process, for clustering algorithms based on center point and distribution, specify the number of clusters (such as 2, 3, 4 ... 30) in order to obtain multiple clustering results; for clustering based on connectivity In the algorithm, the tree structure obtained by clustering is traversed from the root node down layer by layer, and each layer corresponds to a clustering result; for the density-based clustering algorithm, the neighborhood radius is sequentially increased to obtain multiple clusters result. For multiple clustering results obtained using any of the above methods, use clustering evaluation indicators based on the characteristics of the data, such as: Dunn index, Davies–Bouldin index, and Silhouette coefficient (contour coefficient) to score the clustering results To select the best clustering result based on the score.

S4: According to the optimal clustering result, each spatial unit and its corresponding clustering number are obtained as the result of automatic zoning of urban spatial form. The results of the zoning can be exported or displayed as corresponding zoning images of the urban space.

Using the automatic zoning method of urban space morphology according to the embodiment of the present invention, the characteristics of urban space morphology can be understood rationally and comprehensively, and then the result of automatic zoning can be formed, which avoids problems such as subjective judgment or single-dimension dimensionality that are prone to occur in traditional methods. Provide a scientific basis for better understanding and organizing complex systems of urban spatial morphology.

The technical solution of the present invention will be described in detail below by taking the urban space form division of Dengfeng as an example. The specific automatic division method of urban space form mainly includes:

(1) Taking Dengfeng's urban spatial form as the research object, obtain the basic data of its urban spatial form. In this example, the urban spatial morphology raster image is vectorized and divided into spatial units; specifically including:

(1.1) Insert the raster image of the urban space morphology corresponding to the research scope into the CAD software and adjust it to the actual size;

(1.2) Outline the outlines of all blocks in the urban spatial form to form a vectorized spatial unit;

(1.3) Draw the outlines of all buildings inside each block, and mark the number of layers of each building inside the building outline.

(2) Number the divided spatial units, and calculate the index of multiple urban spatial forms for each spatial unit to generate a summary table of urban spatial form characteristics; specifically including:

(2.1) Number the space units composed of

blocks

1,2, ..., n, and calculate 7 indicators for each space unit, the corresponding codes are YDMJ, JZMD, RJL, ZDJZGD, ZDJZMJ, PJJZGD, CLD;

Table 1 Urban Spatial Form Index

(2.2) Summarize the calculation results of YDMJ, JZMD, RJL, ZDJZGD, ZDJZMJ, PJJZGD, and CLD, and generate a summary table of urban spatial morphological characteristics, which is regarded as an n * 7 matrix.

(3) Perform the K-means clustering algorithm based on Mahalanobis distance for the matrix corresponding to the Dengfeng urban spatial morphological characteristics summary table. During the clustering process, set k = 2, 3, ..., 30 in sequence, as shown in Figure 2 As shown in the figure, only the results corresponding to k = 2, 4, 6, and 10 are listed in the figure; for each k value, the Silhouette coefficient is used to score the clustering result, and the highest k value is the best k value. The corresponding clustering result is the optimal clustering result, as shown in FIG. 3, it is shown that when k = 4, it is the optimal clustering result.

For any data point i in the data set, let a (i) denote the average distance from data point i to similar data points, and b (i) denote the minimum value of the average distance from data point i to other types of data points. s (i) represents the silhouette coefficient of data point i. The silhouette coefficient can be defined as:

Note: For the cluster with s (i) = 1, the score is 0. Add this constraint to prevent the number of clusters from increasing significantly. For a data set, its Silhouette coefficient is the average of all data points. The closer the data of the same category is, and the further the distance of the data of different categories, the higher the score. For s (i) close to 1, we need a (i) << b (i). Since s (i) is a measure of how different i is from its own cluster, a smaller value means it matches well. In addition, a large b (i) means that i closely matches its neighboring clusters. Therefore, s (i) close to 1 means that the data is properly clustered. If s (i) is close to a negative value, then through the same logic, we see that it is more appropriate if it is clustered in its adjacent clusters. s (i) close to zero means that the data is located on the boundary of two natural clusters.

(4) According to the optimal clustering result, each spatial unit and its corresponding clustering number are derived to obtain the result of automatic zoning of urban space morphology. This includes:

(4.1) According to the optimal clustering result, each spatial unit and its corresponding clustering number are derived, and the result of automatic zoning of urban spatial form is obtained;

(4.2) Fill the space cells of the same cluster number with one color, and restore the result of the urban space shape partition to the space image, that is, display the result of the partition as a plane geometric image of the colored block.

As shown in FIG. 4, an automatic urban space morphing method disclosed in another embodiment of the present invention further includes:

S5: Analyze the correlation between the urban spatial morphology indicators in the given range, and output the key impact indicators of the automatic division of urban spatial morphology. This includes:

(5.1) Perform the Pearson Correlation Matrix operation on the data in the urban spatial morphological characteristics summary table to generate a correlation matrix diagram, as shown in Figure 5;

(5.2) Comparing the index with the largest sum of absolute values in the correlation matrix graph, in this embodiment, it is ZDJZGD as the key impact index in the spatial form of Dengfeng City.

In addition to Pearson Correlation Matrix, the correlation calculation method used in this step also includes Spearman's Correlation, Kendall Rank Correlation, Goodman and Kruskal's Rank Correlation, Somers' D analysis, etc.

An automatic urban space morphing system disclosed in another embodiment of the present invention includes: a space unit acquisition module for acquiring basic data of urban space morphology within a given range. The basic data of urban space morphology includes or can extract multiple Space unit, the outline of the space unit is a closed polygonal space, including at least one building inside, the outline of the building is a closed polygonal space, the building has layer and / or height information; the space shape index calculation module is used to For each spatial unit, it automatically calculates according to the set of multiple urban spatial morphological indicators to generate a summary table of urban spatial morphological characteristics; where the urban spatial morphological indicators include land area, building density, floor area ratio, maximum base area, maximum building height, At least two of the maximum building area, average building height, and staggering degree; clustering module, used to cluster the matrix corresponding to the urban space morphology feature table using unsupervised clustering algorithm. During the clustering process, the settings are different Clustering parameters for multiple clustering operations, and use based on The cluster evaluation index of the data's own characteristics scores each cluster result to determine the best cluster result; and, the result output module is used to obtain each spatial unit and its corresponding cluster according to the best cluster result Number, as a result of automatic zoning of urban space patterns. Further, the system also includes a key module for generating spatial shape key impact indicators, which is used to analyze the correlation between the urban spatial shape indexes within the scope of the study and obtain the key impact indicators for the urban spatial shape partition. The result output module outputs the results of the automatic zoning of urban space morphology and the key impact indicators of urban space morphology zoning. The system embodiment and the above method embodiment belong to the same inventive concept. For specific implementation details, reference may be made to the above method embodiment, which will not be repeated here.

Based on the same inventive concept, an embodiment of the present invention also discloses an automatic zoning system for urban space morphology. The system includes at least one computer device. The computer device includes a memory, a processor, and is stored on and in the memory. A computer program that is running, and when the computer program is loaded into the processor, the above-mentioned automatic partition method of the urban space form is realized. The parts that are not described in detail in the embodiments of the present invention are all prior art.

Claims

An automatic zoning method for urban space morphology is characterized by the following steps:

(1) Obtain basic data of urban spatial morphology within a given range. The basic data of urban spatial morphology includes or is capable of extracting a plurality of spatial units. The outline of the spatial unit is a closed polygonal space, and the interior includes at least one building. Objects, the outline of the building is a closed polygonal space, and the building has layer and / or height information;

(2) For each spatial unit, automatically calculate according to the set of multiple urban spatial morphological indicators to generate a summary table of urban spatial morphological characteristics; where the urban spatial morphological indicators include land area, building density, floor area ratio, maximum base area, maximum At least two of building height, maximum building area, average building height and staggering degree;

(3) The matrix corresponding to the summary table of urban spatial morphological characteristics is clustered using an unsupervised clustering algorithm. During the clustering process, different clustering parameters are set to perform multiple clustering operations, and clustering based on the characteristics of the data is used. Class evaluation indicators score each clustering result to determine the best clustering result;

(4) According to the optimal clustering result, each spatial unit and its corresponding cluster number are obtained as the result of automatic zoning of urban spatial form.
An automatic urban space morphing method according to claim 1, wherein the method further comprises: performing a correlation analysis on the data in the urban space morphology characteristic summary table, and outputting the key influence of the automatic urban space morphology zoning index.
An automatic urban space morphing method according to claim 1, characterized in that the space unit is a vector space unit formed by dividing blocks, land, or property rights.
An automatic urban space morphing method according to claim 1, wherein the space area of the space unit is obtained by geometrically calculating the polygon space formed by closed polylines; the building density is within the space unit The ratio of the sum of the base areas of all buildings to the space area of the space unit; the floor area ratio is the ratio of the sum of the products of the base area of all buildings in the space unit and the number of building floors and the space area of the space unit; the maximum The base area is the maximum value of the bottom area of all buildings in the space unit; the maximum building height is the maximum value of the height of all buildings in the space unit; the maximum building area is the maximum value of all buildings in the space unit The maximum value of the product of the base area and the number of building layers; the average building height is the average of the heights of all buildings in the space unit; the stagger is the variance of the heights of all buildings in the space unit; The bottom area of the building is obtained by geometric calculation of the polygon space formed by closed polylines.
An automatic partitioning method for urban space morphology according to claim 1, wherein the unsupervised clustering algorithm is a K-means clustering algorithm based on a center point, a hierarchical clustering algorithm based on connectivity, and a density-based clustering algorithm One or more of the DBSCAN clustering algorithm and the distribution-based GMM-EM clustering algorithm; the clustering evaluation index is one of the Dunn index, Davies–Bouldin index, and Silhouette coefficient.
The method for automatically partitioning urban space morphology according to claim 1, wherein the step (3) uses a K-means clustering algorithm based on Mahalanobis distance to the matrix corresponding to the urban space morphology feature summary table Perform clustering. During the clustering process, set different numbers of clusters to perform multiple clustering operations, and use the Silhouette coefficient score to determine the best clustering result; the calculation method of the Silhouette coefficient is:

For any point i in the data set, a (i) represents the average distance from data point i to similar data points, b (i) represents the minimum value of the average distance from data point i to other types of data points, s (i) Represents the silhouette coefficient of data point i; averaging the silhouette coefficients of all data points is the overall silhouette coefficient of the clustering result.
An automatic urban space morphing method according to claim 2, wherein the step of determining the key impact indicators includes:

Perform a correlation operation on the data in the urban spatial morphological characteristics summary table to obtain a correlation matrix diagram;

The index with the largest sum of absolute values in the correlation matrix chart is used as the key impact indicator in the urban spatial form.
An automatic urban space morphing method according to claim 1, wherein in step (4), after obtaining each spatial unit and its corresponding cluster number, the space of the same cluster number The unit is filled with a color, and the results of the urban space morphing partition are displayed as plane geometric images of colored blocks.
An automatic zoning system for urban space morphology, which is characterized by:

The spatial unit acquisition module is used to acquire basic data of the urban spatial form within a given range. The basic data of the urban spatial form includes or is capable of extracting a plurality of spatial units. The outline of the spatial unit is a closed polygonal space. At least one building is included, the outline of the building is a closed polygonal space, and the building is provided with layer and / or height information;

The spatial morphology index calculation module is used to automatically calculate according to the set multiple urban spatial morphology indexes for each spatial unit to generate a summary table of urban spatial morphology characteristics; among them, the urban spatial morphology indexes include land area, building density, floor area ratio, At least two of the maximum base area, maximum building height, maximum building area, average building height, and staggered degree;

The cluster evaluation module is used to cluster the matrix corresponding to the urban space morphological characteristics summary table by unsupervised clustering algorithm. During the clustering process, different clustering parameters are set for multiple clustering operations, and the data-based The cluster evaluation index of its own characteristics scores each cluster result to determine the best cluster result;

And, the result output module is used to obtain each spatial unit and its corresponding cluster number according to the optimal clustering result as the result of the automatic division of the urban spatial form; The key impact indicators of the urban spatial morphology partition are output; the key impact indicators are obtained by performing correlation analysis on the data in the urban spatial morphological characteristics summary table.
An automatic zoning system for urban space morphology includes at least one computer device. The computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. The computer program is characterized by The method for automatically partitioning the urban space form according to any one of claims 1-8 is realized when loading to the processor.