A method and storage medium for identifying urban functional areas based on POI and AOI data

Abstract

The present invention discloses a method and storage medium for identifying urban functional areas based on POI and AOI data, comprising the following steps: dividing the urban space into different block units; obtaining the POI and AOI data of the area; performing coordinate conversion and classification on the POI and AOI data; performing topological processing on the AOI, performing kernel density estimation on the POI according to the category, and obtaining a kernel density distribution map; superimposing the POI, AOI and POI kernel density distribution with the interception unit range respectively, and counting the number of various POIs, AOI area and POI kernel density in the block unit; dividing the block unit into five categories according to the coverage of POI and AOI; identifying the urban functional area according to the five types of blocks combined with the thresholds occupied by the types of POI and AOI in the blocks. The present invention proposes a method for identifying urban functional areas by integrating POI and AOI data, which can be used to identify single or mixed urban functions.

Description

Technical Field

[0001] The present invention relates to the field of geographic space data application technology, and in particular to a method and storage medium for identifying urban functional areas based on POI and AOI data. Background Art

[0002] Urban functional areas are the basic units of urban space and the carriers of social and economic activities. Objectively, quickly and accurately identifying urban functional areas is of great guiding significance for understanding urban spatial structure, rationally planning national space, grasping urban development trends and promoting social and economic development.

[0003] Traditional methods mainly identify urban functional areas through survey statistics and expert judgment. This method is time-consuming and laborious, and has problems such as high subjectivity and poor timeliness. POI (Point of interest) data has the characteristics of large amount of information, high accuracy, simple acquisition method, and fast update cycle. In recent years, it has been increasingly used in the research of urban functional area identification.

[0004] At present, there are four main methods for identifying urban functional areas using POIs. 1) Frequency density method: Based on the number of POIs, frequency density and type proportion are constructed for identification. 2) Weighted method: Public awareness and general area are selected as influencing factors, and POIs are assigned values ​​through weighted method for identification. 3) Reference area method: Identification is performed by weighting the number of POIs and reference area. 4) Kernel density ratio method: Based on POI kernel density, kernel density ratio is used for identification. These methods all have certain problems. The first three methods have the problem that there are no POIs in the research unit and the unit function cannot be identified; the second and third methods have the problem of strong subjectivity due to area assignment; the fourth method has the problem that the kernel density bandwidth affects the recognition accuracy.

[0005] In addition, POI data ignores the actual size of physical objects and is uniformly abstracted as points on the map, which has inherent defects in identifying urban functional areas. AOI (Area of ​​interest) adds the spatial attributes of physical objects based on the original advantages of POI data, so it has obvious advantages over POI in identifying urban functional areas. However, since the number and spatial distribution of AOI are significantly less than POI, it is necessary to combine POI and AOI data to jointly identify urban functional areas. Summary of the invention

[0006] The purpose of the present invention is to provide a method for identifying urban functional areas based on POI and AOI data to solve the problems raised in the above background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for identifying urban functional areas based on POI and AOI data, comprising the following steps:

[0008] Divide the urban space of the object to be identified into different block units;

[0009] Obtain POI and AOI data of the area through electronic maps;

[0010] Perform coordinate conversion and classification on POI and AOI data;

[0011] Perform topological processing on AOI, perform kernel density estimation on POI according to categories, and obtain a kernel density distribution map;

[0012] Overlay the distribution of POI, AOI and POI kernel density with the block unit range, and count the number of POIs, AOI area and POI kernel density in each block unit;

[0013] According to the coverage of POI and AOI, the block units are divided into the following five categories: AOI completely covers the block unit, the block unit has no AOI and has POI, the block unit has no AOI and no POI, AOI covers part of the block unit and the rest has POI, AOI covers part of the block unit and the rest has no POI;

[0014] Identify urban functional areas based on the type proportions of POIs and AOIs in five types of block units.

[0015] Preferably, the "AOI fully covered block unit" is defined as the first type of block unit, and its functional weight model is:

[0016]

[0017] Among them, a i is the AOI area of ​​the i-th function in the block; S is the block area; A i is the proportion of the AOI area of ​​the i-th function to the block area.

[0018] Preferably, "block unit without AOI but with POI" is defined as the second type of block unit, and its functional weight model includes:

[0019] Calculate POI frequency density:

[0020]

[0021] Among them, n i is the number of POIs with the i-th function in the block; N i is the total number of POIs with the i-th function; F iis the frequency density of the i-th functional POI in the total number of functional POIs;

[0022] Calculate the proportion of POI types:

[0023]

[0024] Among them, C i is the proportion of the frequency density of the ith functional POI to the frequency density of all POIs in the block.

[0025] Preferably, "block unit without AOI and POI" is defined as the third type of block unit, and its functional proportion model is:

[0026]

[0027] Among them, d i is the sum of the kernel densities of the POIs with the i-th function in the block; K is the total number of kernel densities in the block; D i is the proportion of the kernel density of the POI with the i-th function to the total kernel density.

[0028] Preferably, "AOI covers part of the block unit and the rest" contains POI, which is defined as the fourth type of block unit, and its functional proportion model is:

[0029] AC i =A i ×P+C i ×(1-P),i∈[1,6]

[0030] Among them, P is the proportion of all functional AOI areas in the block to the block area, A i is the proportion of the AOI area of ​​the i-th function to the block area, C i is the proportion of the frequency density of the POI with the i-th function to the frequency density of all POIs in the block, AC i is the proportion of the ith function.

[0031] Preferably, "AOI covers part of the block unit and there is no POI in the rest" is defined as the fifth type of block unit, and its functional proportion model is:

[0032] AD i =A i ×P+D i ×(1-P),i∈[1,6]

[0033] Among them, P is the proportion of all functional AOI areas in the block to the block area; A i is the proportion of the AOI area of ​​the i-th function to the block area, D i is the proportion of the kernel density of the POI with the i-th function to the total kernel density, AD i is the proportion of the ith function.

[0034] Preferably, urban functions are divided into 6 categories, and POI and AOI are also classified accordingly, namely: public administration and public service land, commercial service land, residential land, industrial and mining land, green land and open space land, and transportation land.

[0035] Preferably, the principle of urban function identification is to set a threshold value according to the proportion of each type of urban function in the block unit. If the proportion of a certain urban function is greater than or equal to the threshold, then the urban function is the main function of the block unit, otherwise, the block unit has mixed functions.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention proposes a method for identifying urban functional areas by integrating POI and AOI data, which can be used to identify single or mixed urban functions. At the same time, the coverage of POI and AOI data is creatively sorted out into five types of block units. Through the separate analysis and identification of the five types of block units, the accuracy, coverage and objectivity are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flowchart of the specific technology implementation of the present invention;

[0039] Figure 2 POI distribution map used in the implementation case of the present invention;

[0040] Figure 3 AOI distribution diagram used in the implementation case of the present invention;

[0041] Figure 4 This is the POI kernel density distribution map, (a) represents public administration and public service land, (b) represents commercial service land, (c) represents residential land, (d) represents industrial and mining land, (e) represents green land and open space land; (f) represents transportation land;

[0042] Figure 5 This is a classification result map of urban functional areas obtained by using the method of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] This embodiment takes Shinan District, Shibei District and Licang District (referred to as "three districts") of Qingdao City as an example to illustrate the specific implementation of the present invention. The specific implementation steps of the present invention for identifying urban functional areas will be specifically described below in conjunction with this example:

[0045] (1) Road network processing

[0046] 1) Download OSM road network data and extract main roads and secondary roads.

[0047] 2) Convert the main road network data into the CGCS2000 coordinate system.

[0048] 3) Organize the main road network data and delete the road networks that cannot be effectively connected to other road networks.

[0049] 4) Perform a “line-to-surface” operation on the main road network data and divide the three districts into 291 block units as the basic units for identifying urban functional areas.

[0050] (2) POI processing

[0051] 1) Use Baidu Map API to obtain POI data within the three zones. The main attributes include location information (longitude, latitude, address), category, name, etc. Baidu Map is divided into 21 categories: food, hotels, shopping, life services, beauty, tourist attractions, leisure and entertainment, sports and fitness, education and training, culture and media, medical care, car services, transportation facilities, finance, real estate, companies and enterprises, government agencies, entrances and exits, natural features, administrative landmarks, and gate addresses.

[0052] 2) The POI is converted to the Mars coordinate system, WGS84, and CGCS2000 in turn.

[0053] 3) Since some categories of POI (such as gate addresses, entrances and exits, administrative landmarks, etc.) have no actual urban function representation meaning in national land space planning, these categories of POI are deleted. Since logistics companies in POIs are commercial service facilities and rarely occupy independent land, they are inconsistent with the meaning of warehousing land; there is no POI corresponding to public utility land; refer to the "Guidelines for Classification of Land and Sea Use for National Land Space Survey, Planning, and Use Control (Trial)", 6 categories of POI are extracted.

[0054] 4) Perform kernel density estimation on the six types of POIs. When estimating kernel density, the bandwidth has a significant impact on the results. After multiple tests, it was found that 300 meters is the optimal bandwidth for kernel density estimation of POIs in three zones.

[0055] (3) AOI processing

[0056] 1) Use Baidu Map API to obtain AOI data within the three zones. Compared with POI, AOI not only has location information (longitude, latitude, address), category, name and other attributes, but also has spatial boundary attributes. In addition, AOI classification is consistent with POI.

[0057] 2) The AOI data is converted to the Mars coordinate system, WGS84, and CGCS2000 in sequence.

[0058] 3) Similar to the POI operation, delete the AOIs that have no significance in representing urban functions; extract 6 types of AOIs.

[0059] 4) Perform topological processing on AOI data, and merge, cut, and process AOIs with topological errors by referring to street view images and high-precision remote sensing images.

[0060] (4) Block unit overlay analysis

[0061] 1) Overlay analysis of AOI and block units, and statistics of AOI area in each block unit.

[0062] 2) AOI and block unit erasure analysis to obtain AOI-free block units.

[0063] 3) Overlay analysis of POIs and blocks without AOIs, and the number of various POIs in the statistical unit.

[0064] 4) Overlay analysis of POI kernel density and AOI-free block units, and statistical analysis of the kernel density of various POIs in the units.

[0065] (5) Identification of urban functional areas

[0066] 1) According to the coverage of POI and AOI, the block units are divided into five categories, namely, AOI completely covers the block unit (i.e., the first category of block units), the block unit has no AOI and has POI (i.e., the second category of block units), the block unit has no AOI and no POI (i.e., the third category of block units), AOI covers part of the block unit and there is POI in the rest of the block unit (i.e., the fourth category of block units), and AOI covers part of the block unit and there is no POI in the rest of the block unit (i.e., the fifth category of block units).

[0067] 2) Calculation of urban functions of the first type of block units

[0068] The urban function proportion model is:

[0069]

[0070] Among them, a i is the AOI area of ​​the i-th function in the block; S is the block area; A i is the proportion of the AOI area of ​​the i-th function to the block area.

[0071] 3) Calculation of urban functions of the second type of block units

[0072] The urban function proportion model includes:

[0073] a. Calculate POI frequency density:

[0074]

[0075] Among them, n i is the number of POIs with the i-th function in the block; N i is the total number of POIs with the i-th function; F i is the frequency density of the i-th functional POI to the total number of POIs with this function.

[0076] b. Calculate the proportion of POI types:

[0077]

[0078] Among them, C i is the proportion of the frequency density of the ith functional POI to the frequency density of all POIs in the block.

[0079] 4) Calculation of urban functions of the third type of block unit

[0080] The proportion model of various urban functions is:

[0081]

[0082] Among them, d i is the sum of the kernel densities of the POIs with the i-th function in the block; K is the total number of kernel densities in the block; D i is the proportion of the POI core density of the i-th function to the total block core density.

[0083] 5) Calculation of urban functions of the fourth type of block unit

[0084] The proportion model of various urban functions is:

[0085] AC i =A i ×P+C i ×(1-P),i∈[1,6]

[0086] Among them, P is the proportion of all functional AOI areas in the block to the block area; AC i is the proportion of the ith function.

[0087] 6) Calculation of urban functions of the fifth type of block unit

[0088] The proportion model of various urban functions is:

[0089] AD i =Ai ×P+D i ×(1-P),i∈[1,6]

[0090] Among them, P is the proportion of all functional AOI areas in the block to the block area; AD i is the proportion of the ith function.

[0091] 7) Identification of urban functional areas

[0092] If the proportion of a certain type of urban function in a block unit exceeds 55%, then this urban function is regarded as the main function of the block unit. If the proportion of each type of urban function in a block unit is less than 55%, then the block unit is mixed function.

[0093] 8) Analysis of the accuracy of urban functional area identification

[0094] The method of the present invention, the frequency density method, and the kernel density ratio method were used to identify the urban function areas of the three districts, and the identification results were compared with the results of manual visual interpretation and the accuracy was evaluated. It was found that the accuracy of the method of the present invention reached 82.03%, which was significantly higher than the other two methods. In addition, the method of the present invention can identify the urban functions of 291 block units in the three districts, and the identification coverage rate is higher than that of the frequency density method.

[0095] Table 1 Comparison of recognition results

[0096] Method of the present invention Frequency Density Method Nuclear Density Gravimetry Coverage 100％ 99.31％ 100％ accuracy 82.03％ 73.62％ 67.75％

[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for identifying urban functional areas based on POI and AOI data, characterized in that: The steps include: Divide the urban space of the object to be identified into different block units; Obtain POI and AOI data of the area through electronic maps; Perform coordinate conversion and classification on POI and AOI data; Perform topological processing on AOI, perform kernel density estimation on POI according to categories, and obtain a kernel density distribution map; Overlay the distribution of POI, AOI and POI kernel density with the block unit range, and count the number of POIs, AOI area and POI kernel density in each block unit; According to the coverage of POI and AOI, the block units are divided into the following five categories: AOI completely covers the block unit, the block unit has no AOI and has POI, the block unit has no AOI and no POI, AOI covers part of the block unit and the rest has POI, AOI covers part of the block unit and the rest has no POI; The "AOI fully covered block unit" is defined as the first type of block unit, and its functional proportion model is: in, a i For the block i Functional AOI area; S is the block area; A i For the i The proportion of AOI area of ​​each function to the block area; The "block unit without AOI but with POI" is defined as the second type of block unit, and its functional proportion model includes: Calculate POI frequency density: in, n i For the block i Number of POIs with different functions; N i For the i Total number of functional POIs; F i For the i The frequency density of a functional POI to the total number of functional POIs; Calculate the proportion of POI types: in, C i For the i The frequency density of POI with certain functions accounts for the proportion of the frequency density of all POIs in the block; The "block unit without AOI and POI" is defined as the third type of block unit, and its functional proportion model is: in, d i For the block i The sum of the kernel densities of various functional POIs; K is the total number of block core densities; D i For the i The proportion of the functional POI kernel density to the total kernel density; The "AOI covers part of the block unit and the rest has POI" is defined as the fourth type of block unit, and its functional proportion model is: in, P is the proportion of all functional AOI areas in the block to the block area, A i For the i The proportion of functional AOI area to the block area, C i For the i The frequency density of POIs with different functions accounts for the proportion of all POI frequency densities in the block. AC i For the i Functional proportion; The "AOI covers part of the block unit and there is no POI in the rest" is defined as the fifth type of block unit, and its functional proportion model is: in, P It is the ratio of all functional AOI areas in the block to the block area; A i For the i The proportion of functional AOI area to the block area, D i For the i The proportion of the functional POI kernel density to the total kernel density, AD i For the i Functional proportion; Identify urban functional areas based on the proportion of POI and / or AOI types in five types of block units; The urban function identification principle is to set a threshold value according to the proportion of various urban functions in the block unit. If the proportion of a certain urban function is greater than or equal to the threshold, then the urban function is the main function of the block unit, otherwise, the block unit has mixed functions.

2. The method for identifying urban functional areas based on POI and AOI data according to claim 1, characterized in that: Urban functions are divided into 6 categories, and POI and AOI are also classified according to them, namely: public administration and public service land, commercial service land, residential land, industrial and mining land, green space and open space land, and transportation land.

3. A storage medium, characterized in that: A computer program is stored, which runs a method for identifying urban functional areas based on POI and AOI data as described in any one of claims 1 to 2.

Images