Multi-level building classification method and system based on multi-source space big data

Abstract

The invention discloses a multi-level building classification method and system based on multi-source space big data, and relates to the technical field of urban building classification, the specific steps are as follows: obtaining building vector data, traffic network data and POI point data, the POI point data including first POI point data and second POI point data; generating urban plot vector data based on the traffic network data; based on a first mapping relation, utilizing the first POI point data to generate urban land parcel classification data; based on a second mapping relation, building function classification data are generated by utilizing the second POI point data; and according to the urban land parcel classification data and the building function classification data, carrying out two times of assignment on building function types based on the building vector data and the urban land parcel vector data. According to the invention, the accuracy of building classification is effectively improved, and more comprehensive and accurate analysis can be carried out on urban spatial data.

Description

Technical Field

[0001] The present invention relates to the technical field of urban building classification, and in particular to a multi-level building classification method based on multi-source spatial big data. Background Art

[0002] At present, there are many challenges in the accuracy of building classification methods based on spatial big data. Since the original spatial data lacks unified standards for classification and has a certain degree of randomness in geographic spatial distribution, there is a large uncertainty when used to identify building categories. This makes it difficult for current building classification methods to meet the urgent need to quickly realize large-scale building classification using spatial big data in practice.

[0003] With the continuous development of spatial big data in the field of smart city construction, higher requirements are put forward for refined urban planning and management. Among them, POI point data, as information representing the specific location of geographic entities, has been widely used in the identification research of different functional areas in cities due to its semantic richness and high timeliness. With the characteristics of low acquisition cost and fast update, POI point data has become an effective auxiliary tool for decision makers in population surveys, land management, energy conservation and emission reduction, business analysis and other businesses, and is of great significance to improving the efficiency of urban management. For example, the invention patent with application publication number CN110008542A and name of a method for classifying massive urban buildings uses public data sources to obtain the building base plan and POI, and spatially connects the two in Arcmap. A certain number of training samples are selected, and a logistic regression model is used to establish a classifier consistent with the number of target classifications. The parameter weights are iteratively obtained through Python language programming. Next, a certain number of verification samples are selected, and the parameter weights are substituted into each classifier to obtain the value, and the type corresponding to the classifier with the largest value is selected as the predicted type of the building. For example, in the invention patent with application publication number CN107247938A and titled "A method for functional classification of urban buildings in high-resolution remote sensing images", the CNN method is used to extract buildings from high-resolution remote sensing images to obtain building extraction results; the POI data is sorted and classified according to attribute information, and the kernel density of the POIs of commercial service facilities land, public management and public service land, and residential land is estimated respectively, and the kernel density maps of these land types are obtained respectively; the kernel density average of a single building is calculated using the above-mentioned CNN-based remote sensing image building information extraction results and kernel density maps. It can be seen that the connection between the building base plan and POI data in CN110008542A is completely based on the rectangular range of the building base, and the building function is assigned according to the kernel density map of POI points falling within the building vector range in CN107247938A. However, due to the uncertainty of the geographic spatial distribution of POI points, in many cases the POI points related to the building object are not located within the rectangular range of the building vector data, resulting in the loss of key information and directly affecting the accuracy of the building function classification. At the same time, both CN110008542A and CN107247938A do not distinguish between POI points with different distribution characteristics, ignoring the impact of the differences between high-density and low-density POI points in the building classification process, limiting the applicable scope of the building classification method and reducing the feasibility in practical applications. Therefore, how to solve the above problems is a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0004] The present invention provides a multi-level building classification method and system based on multi-source spatial big data to solve the problems existing in the above-mentioned prior art.

[0005] The present invention first provides a multi-level building classification method based on multi-source spatial big data, the specific steps are:

[0006] Acquire building vector data, traffic network data and POI point data, wherein the POI point data includes first POI point data and second POI point data;

[0007] Generate urban land parcel vector data based on the traffic road network data;

[0008] Based on a first mapping relationship, the first POI point data is used to generate urban plot classification data; based on a second mapping relationship, the second POI point data is used to generate building function classification data; the first mapping relationship is a mapping relationship between the first POI point data and the urban plot type; the second mapping relationship is a mapping relationship between the second POI point data and the building function type;

[0009] According to the urban plot classification data and the building function classification data, the building function type is assigned twice based on the building vector data and the urban plot vector data.

[0010] Optionally, the steps for generating the urban land parcel vector data are:

[0011] Obtain traffic road network data and traffic road grade data;

[0012] Adding a buffer zone based on the traffic road network data and the traffic road grade data;

[0013] The traffic land is determined according to the buffer zone, and the vector data of the urban land parcel is obtained by using the intersection inversion method.

[0014] Optionally, the step of acquiring the first mapping relationship is:

[0015] Acquire the POI point data of any city, and extract the first POI point data from the POI point data of any city;

[0016] Classify urban plots according to the first POI point data of any city;

[0017] According to the mapping relationship between the city plot type and the first POI point data, the city plot type field is correspondingly added to the first POI point data to obtain the first mapping relationship.

[0018] Optionally, the second mapping relationship is obtained by:

[0019] Acquire the POI point data of any city, and extract the second POI point data from the POI point data of any city;

[0020] Acquire the spatial position of the second POI point data, and acquire the urban plot type of the urban plot to which the second POI point data belongs according to the spatial position of the second POI point data;

[0021] Classify the buildings according to the second POI point data to obtain target categories;

[0022] Determine the building function type based on the second POI point data, the urban plot type and the target category;

[0023] According to the mapping relationship between the second POI point data and the building function type and the urban plot type of the urban plot to which the second POI point belongs, the building function type field and the urban plot type field are added to the second POI point data accordingly to obtain the second mapping relationship.

[0024] Optionally, the step of acquiring the second mapping relationship may further include:

[0025] Acquire the POI point data of any city, and extract the second POI point data from the POI point data of any city;

[0026] Acquire the spatial position of the second POI point data, and acquire the urban plot type of the urban plot to which the second POI point data belongs according to the spatial position of the second POI point data;

[0027] Adding a city plot type field to the second POI point data to form third POI point data;

[0028] Classify the buildings according to the second POI point data to obtain target categories;

[0029] Determine the building function type based on the third POI point data and the target category;

[0030] According to the mapping relationship between the third POI point data and the building function type, a building function type field is correspondingly added to the third POI point data to obtain the second mapping relationship.

[0031] Optionally, the specific steps for assigning building function types are:

[0032] Acquire the building vector data, the urban plot vector data, the urban plot classification data and the building function classification data;

[0033] Assigning values ​​to the urban plot vector data using a frequency density ratio method according to the urban plot vector data and the urban plot classification data to obtain urban plot type data;

[0034] According to the urban plot type data and the building function classification data, the building function type is assigned based on the building vector data through a spatial weight assignment algorithm to obtain the building function type.

[0035] Alternatively, the expression for the frequency density ratio method is:

[0036]

[0037] Where: i represents the urban plot type; F i represents the frequency density of the i-th type of POI points in any urban area to the total number of POI points of this type; C i It represents the ratio of the frequency density of the i-th type of POI point in any urban plot to the frequency density of all types of POI points in the urban plot.

[0038] Optionally, the expression of the spatial weight assignment algorithm is:

[0039]

[0040] Where: m represents the number of POI points of the i-th category within the radius range D with the center point of the building object as the center; w j Represents weight; N i Represents the total number of POI points of the i-th category in the POI dataset.

[0041] Optionally, the step of acquiring the POI point data is: acquiring original data, and performing elimination and merging processing on the original data.

[0042] The present invention also provides a multi-level building classification system based on multi-source spatial big data, comprising:

[0043] A data acquisition module, used to acquire building vector data, traffic network data and POI point data, wherein the POI point data includes first POI point data and second POI point data;

[0044] An urban plot division module, used for generating urban plot vector data according to the traffic road network data;

[0045] A POI point classification module, configured to generate urban plot classification data using the first POI point data according to a first mapping relationship; and generate building function classification data using the second POI point data based on a second mapping relationship; the first mapping relationship is a mapping relationship between the first POI point data and the urban plot type; the second mapping relationship is a mapping relationship between the second POI point data and the building function type;

[0046] The building classification module is used to assign building function types twice based on the building vector data and the urban plot vector data according to the urban plot classification data and the building function classification data.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. The present invention provides a standardized processing flow for classifying POI point data based on two dimensions: urban plots and buildings, which effectively improves the accuracy of building classification and can be applied in different scenarios to conduct a more comprehensive and accurate analysis of urban spatial data;

[0049] 2. The present invention introduces the normalization processing of POI point data categories, divides POI points into two categories: high importance and low density and low importance and high density, clarifies the use principles of data with different spatial distribution characteristics in building classification, and makes up for the shortcomings of such data in application standardization; and through the standardized and fine classification of POI points, the pertinence and adaptability of the building classification method are improved, which facilitates the wider application of spatial big data. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without creative work.

[0051] Figure 1 It is a schematic flow chart of the method provided by the present invention;

[0052] Figure 2 It is a schematic diagram of the process of obtaining the urban land parcel vector data in the present invention;

[0053] Figure 3 It is a schematic diagram of the process of the reclassification method based on POI point data in the present invention;

[0054] Figure 4 A schematic diagram of the flow chart of the method for assigning building function types according to the present invention;

[0055] Figure 5 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0056] 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.

[0057] The application principle of the present invention is described in detail below in conjunction with the accompanying drawings.

[0058] The first aspect of the present invention provides a multi-level building classification method based on multi-source spatial big data, such as Figure 1 As shown, the specific steps are:

[0059] Step 1: Acquire building vector data, traffic network data and POI point data, where the POI point data includes first POI point data and second POI point data;

[0060] Step 2: Generate urban land vector data based on traffic network data;

[0061] Step 3: Based on the first mapping relationship, the first POI point data is used to generate urban land parcel classification data; based on the second mapping relationship, the second POI point data is used to generate building function classification data; the first mapping relationship is a mapping relationship between the first POI point data and the urban land parcel type; the second mapping relationship is a mapping relationship between the second POI point data and the building function type;

[0062] Step 4: According to the urban plot classification data and the building function classification data, the building function type is assigned twice based on the building vector data and the urban plot vector data.

[0063] In one embodiment, the steps of acquiring building vector data, traffic network data and POI point data in step 1 are: acquiring multi-source spatial data in a specific area through various data collection methods, and the multi-source spatial data includes three types: building vector data, traffic network data and POI point data. Building vector data is a data type commonly used in geographic information systems (GIS) to represent building geometric information in vector form, usually describing the specific location of the building in the form of linear features and outer contours, including the geometric form, location and attribute information of the building. Traffic network data refers to geographic information data that records the structure and topological relationship of urban transportation networks, usually describing the location, length, link relationship and attribute information of transportation facilities such as roads, streets and railways. POI points are used to represent points of interest or locations of interest on the map, and are used to represent data points with attributes of specific geographical locations, including: basic information: name, address, coordinates; classification information: category / type, such as restaurant, school, store, scenic spot, etc.; sub-category: such as Chinese restaurant, Western restaurant; contact information; business information, etc.; among them, basic information and classification information are more relevant to the POI information referred to in this embodiment. The following methods can be used to obtain relevant spatial data:

[0064] (1) Using public data from government management departments;

[0065] (2) Download it for free using open source platforms such as OpenStreetMap;

[0066] (3) Obtain the required geospatial data based on map platform APIs (Baidu, Amap, etc.).

[0067] In one embodiment, if Figure 2 As shown in the figure, the steps for generating urban plot vector data are:

[0068] Obtain traffic road network data and traffic road grade data;

[0069] Add buffer zones based on traffic network data and traffic road grade data;

[0070] The transportation land is determined according to the buffer zone, and the vector data of urban plots is obtained by using the intersection inversion method.

[0071] Furthermore, as a linear geographic information data, traffic network data can naturally divide urban space into different basic units. Specifically, through the spatial analysis tools in GIS software, buffers are added based on the original traffic network data according to different traffic road levels, and the buffer range represents the corresponding road width. The buffer range based on the road centerline is defined as traffic land. The intersection negation function in spatial analysis is used to process the traffic map layer and the study area range layer to obtain spatial data representing urban plots. Urban plots will be saved and recorded in the geographic information system in the form of vector information.

[0072] In one embodiment, the step of acquiring POI point data is: acquiring original data, and performing elimination and merging processing on the original data.

[0073] Furthermore, due to the large number of data categories in the POI points, some information cannot provide effective information for the classification of building functions, and even causes interference. Therefore, it is necessary to eliminate and merge the original data according to actual needs. According to the spatial distribution characteristics of the POI point data, it is divided into the first POI point data and the second POI point data, where the first POI point data is high-importance and low-density POI point data (residential areas, commercial office buildings, industrial parks, colleges and universities, etc.); the second POI point data is low-importance and high-density POI point data (restaurants, convenience stores, companies, specialty stores, etc.). High-importance and low-density POI point data are used for the reclassification of urban plots, and low-importance and high-density POI point data are used for the reclassification of building types.

[0074] In one embodiment, the step of obtaining the first mapping relationship is:

[0075] Obtain POI point data of any city, and extract first POI point data from the POI point data of any city;

[0076] Classify urban plots according to the first POI point data of any city;

[0077] According to the mapping relationship between the city plot type and the first POI point data, the city plot type field is correspondingly added to the first POI point data to obtain the first mapping relationship.

[0078] Furthermore, the reclassification of POI point data on urban land vector data mainly refers to the national standard "Land Use Classification" (GB / T 21010-2017), and takes the high-importance and low-density POI point categories as the benchmark to classify urban land according to functional types into residential land, commercial and service land, industrial land, education and scientific research land, medical and health land, cultural and sports land, government agency land, transportation facility land, park green land, etc. Figure 3As shown, in the process of reclassification, based on the mapping relationship between the original POI type and the urban plot classification target, the urban plot type field is added to the original POI point attribute table and assigned a value to obtain the first mapping relationship, that is, the urban plot classification attribute table based on the POI point type, as shown in Table 1, which shows the mapping relationship between the specific POI type and the classified urban plot.

[0079] Table 1 Classification attribute table of urban plots based on POI point type

[0080]

[0081]

[0082] In one embodiment, the second mapping relationship is obtained by:

[0083] Obtain POI point data of any city, and extract second POI point data from the POI point data of any city;

[0084] Obtaining the spatial position of the second POI point data, and obtaining the urban plot type of the urban plot to which the second POI point data belongs according to the spatial position of the second POI point data;

[0085] Classify the buildings according to the second POI point data to obtain the target category;

[0086] Determine the building function type based on the second POI point data, urban plot type and target category;

[0087] According to the mapping relationship between the second POI point data and the building function type and the urban plot type of the urban plot to which the second POI point belongs, the building function type field and the urban plot type field are correspondingly added to the second POI point data to obtain a second mapping relationship.

[0088] In one embodiment, the step of acquiring the second mapping relationship may further include:

[0089] Obtaining POI point data of any city, and extracting second POI point data from the POI point data of any city;

[0090] Obtaining the spatial position of the second POI point data, and obtaining the urban plot type of the urban plot to which the second POI point data belongs according to the spatial position of the second POI point data;

[0091] Adding the city plot type field to the second POI point data to form the third POI point data;

[0092] Classify the buildings according to the second POI point data to obtain the target category;

[0093] Determine the building function type based on the third POI point data and target category;

[0094] According to the mapping relationship between the third POI point data and the building function type, the building function type field is correspondingly added to the third POI point data to obtain a second mapping relationship.

[0095] On the basis of urban plot types, it is necessary to further establish a mapping relationship between low-importance and high-density POI point categories and functional categories of building objects. Taking the application of building energy consumption calculation in the low-carbon city development scenario as an example, for the classification of POI point data in terms of building functions, the "Energy Consumption Standard for Civil Buildings" (GB / T 51161-2016) can be referred to, and building objects can be divided into residential buildings, shopping malls, office buildings, hotels, hospitals, schools, cultural and sports venues, factory buildings, government agencies, transportation facilities, etc. First, based on the spatial position of the low-importance and high-density POI point data in the urban plot, it is necessary to add the urban plot type as a new field to the attribute table of the original POI point data. In the reclassification process, the building function type is jointly determined based on the specific POI type and the urban plot type, and the building function type field is added to the attribute table of the original POI point data accordingly to obtain the second mapping relationship, that is, the building function classification attribute table based on the POI point data type, as shown in Table 2, which shows the mapping relationship between the specific type of the original POI point data, the urban plot type to which the POI point data belongs, and the classified building function type.

[0096] Table 2 Building function classification attribute table based on POI point data type

[0097]

[0098]

[0099] In one embodiment, the specific steps of assigning building function types are:

[0100] Obtain building vector data, urban plot vector data, urban plot classification data and building function classification data;

[0101] According to the urban plot vector data and the urban plot classification data, the urban plot vector data is assigned values ​​by using the frequency density ratio method to obtain the urban plot type data;

[0102] According to the urban plot type data and building function classification data, the building function type is assigned on the basis of the building vector data through the spatial weight assignment algorithm to obtain the building function type.

[0103] Furthermore, if Figure 4As shown in the figure, the building function classification is to assign values ​​based on the reclassified POI point data at the two levels of urban plots and building vectors. First, based on the urban plot vector data and the urban plot classification attribute table based on the POI point type, the frequency density ratio method is used to calculate the frequency of different types of POI point data in the urban plot, and the urban plot type with the highest proportion is selected to assign values ​​to the urban plot. Based on the results of urban plot classification, combined with the mapping relationship between POI point data and building function types in the building function classification attribute table based on the POI point data type, the building function type is assigned based on the building vector data through the spatial weight assignment algorithm. This process aims to utilize spatial big data, through the processing of urban plots and buildings at two levels, to realize the classification of building functions more comprehensively and accurately, and provide strong support for the in-depth understanding and management of urban space.

[0104] In one embodiment, each urban plot has a number of POI point data representing various geographic entities. According to the mapping relationship between the POI point data and the functions of the urban plot, for each urban plot, the POI point data falling within the scope of the plot can be screened out according to the spatial scope, and the frequency density and type ratio of indicators are constructed based on the type of the POI point data to judge the function of the urban plot.

[0105]

[0106] Where: i represents the urban plot type; n i Indicates the number of POI points of type i in a certain city plot; N i represents the total number of POI points of the i-th category in the POI dataset; F i represents the frequency density of the i-th type of POI points in a certain city plot to the total number of POI points of this type; C i It represents the ratio of the frequency density of the POI point of the i-th type in a certain city plot to the frequency density of all types of POI points in the city plot. i The POI point category with the largest value is assigned a value.

[0107] In one embodiment, when assigning building functions based on the classification of urban plots, since the spatial range of buildings is relatively small and most POI points are sparsely distributed, there are no POI points within the range of many buildings. The assignment method based on the spatial vector boundary cannot achieve the classification of building function types at this spatial granularity. Therefore, when assigning building functions, it is necessary to expand the range of POI points involved in the calculation, ignore the building vector boundary, and perform density frequency F iDistance weights are assigned to reflect the impact of spatial distance on the attribution of POI points to specific buildings. Taking the inverse distance weighted algorithm as an example, for a certain building object, let D be the boundary radius of its weight assignment, then its density frequency F i The expression of building assignment is shown in formula (3):

[0108]

[0109] Where: m is the number of POI points of the i-th category within the radius range D with the center point of the building object as the center, and the weight w is j The calculation of adopts a one-dimensional Gaussian function with the distance from the POI point to the building as the independent variable. The one-dimensional graph of the Gaussian function is a bell-shaped curve, and its expression is shown in formula (4):

[0110]

[0111] Where a is the peak value, b is the independent variable value at the peak value, and c is the standard deviation. The meaning of the independent variable x is the distance from the POI point to the geometric center of the building plane range polygon. When the POI point is close to the building, the weight should be close to 1. When the POI point is too far from the building, the weight is close to 0 but not negative. Therefore, let a = 1 and b = 0.

[0112] For a specified building object, refer to formula (2) and select the C within the boundary of the building object's spatial weight assignment. i The POI point category with the largest value is assigned a functional category.

[0113] Another aspect of this embodiment includes a multi-level building classification system based on multi-source spatial big data, such as Figure 5 As shown, including:

[0114] A data acquisition module, used to acquire building vector data, traffic network data and POI point data, wherein the POI point data includes first POI point data and second POI point data;

[0115] The urban plot division module is used to generate urban plot vector data based on traffic road network data;

[0116] A POI point classification module is used to generate urban plot classification data using first POI point data according to a first mapping relationship; and to generate building function classification data using second POI point data based on a second mapping relationship; the first mapping relationship is a mapping relationship between the first POI point data and the urban plot type; the second mapping relationship is a mapping relationship between the second POI point data and the building function type;

[0117] The building classification module is used to assign building function types twice based on building vector data and urban plot vector data according to urban plot classification data and building function classification data.

[0118] Furthermore, the data acquisition module collects multi-source spatial big data, summarizes and organizes them according to their categories, and then inputs the pre-processed spatial data into the urban plot division module, the POI point classification module and the building classification module respectively. The urban plot division module divides the main areas in the city based on the traffic network information, identifies the basic units of the land use pattern in the area, and inputs the divided plots into the building classification module. The POI point classification module reclassifies according to the POI point data attributes, and then inputs the classification results into the building classification module. Finally, in the building classification module, the building is functionally classified based on the spatial range of the POI point and the building vector rectangle. This integrated system enables the modules to cooperate with each other, realizing the comprehensive classification and accurate function identification of urban spatial data. In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0119] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-level building classification method based on multi-source spatial big data, characterized in that: The specific steps are: Acquire building vector data, traffic network data and POI point data, wherein the POI point data includes first POI point data and second POI point data; Generate urban land parcel vector data based on the traffic road network data; Based on the first mapping relationship, generating urban land parcel classification data using the first POI point data; Based on the second mapping relationship, generating building function classification data using the second POI point data; The first mapping relationship is a mapping relationship between the first POI point data and the urban plot type; the second mapping relationship is a mapping relationship between the second POI point data and the building function type; According to the urban plot classification data and the building function classification data, the building function type is assigned twice based on the building vector data and the urban plot vector data.

2. The multi-level building classification method based on multi-source spatial big data according to claim 1 is characterized in that: The steps for generating the urban plot vector data are as follows: Obtain traffic road network data and traffic road grade data; Adding a buffer zone based on the traffic road network data and the traffic road grade data; The traffic land is determined according to the buffer zone, and the vector data of the urban land parcel is obtained by using the intersection inversion method.

3. The multi-level building classification method based on multi-source spatial big data according to claim 1 is characterized in that: The steps of obtaining the first mapping relationship are: Acquire the POI point data of any city, and extract the first POI point data from the POI point data of any city; Classify urban plots according to the first POI point data of any city; According to the mapping relationship between the city plot type and the first POI point data, the city plot type field is correspondingly added to the first POI point data to obtain the first mapping relationship.

4. The multi-level building classification method based on multi-source spatial big data according to claim 3 is characterized in that: The second mapping relationship is obtained by: Acquire the POI point data of any city, and extract the second POI point data from the POI point data of any city; Acquire the spatial position of the second POI point data, and acquire the urban plot type of the urban plot to which the second POI point data belongs according to the spatial position of the second POI point data; Classify the buildings according to the second POI point data to obtain target categories; Determine the building function type based on the second POI point data, the urban plot type and the target category; According to the mapping relationship between the second POI point data and the building function type and the urban plot type of the urban plot to which the second POI point belongs, the building function type field and the urban plot type field are added to the second POI point data accordingly to obtain the second mapping relationship.

5. The multi-level building classification method based on multi-source spatial big data according to claim 3 is characterized in that: The step of acquiring the second mapping relationship may further include: Acquire the POI point data of any city, and extract the second POI point data from the POI point data of any city; Acquire the spatial position of the second POI point data, and acquire the urban plot type of the urban plot to which the second POI point data belongs according to the spatial position of the second POI point data; Adding a city plot type field to the second POI point data to form third POI point data; Classify the buildings according to the second POI point data to obtain target categories; Determine the building function type based on the third POI point data and the target category; According to the mapping relationship between the third POI point data and the building function type, a building function type field is correspondingly added to the third POI point data to obtain the second mapping relationship.

6. The multi-level building classification method based on multi-source spatial big data according to claim 1 is characterized in that: The specific steps for assigning building function types are: Acquire the building vector data, the urban plot vector data, the urban plot classification data and the building function classification data; Assigning values ​​to the urban plot vector data using a frequency density ratio method according to the urban plot vector data and the urban plot classification data to obtain urban plot type data; According to the urban plot type data and the building function classification data, the building function type is assigned based on the building vector data through a spatial weight assignment algorithm to obtain the building function type.

7. The multi-level building classification method based on multi-source spatial big data according to claim 6 is characterized in that: The expression of frequency density ratio method is: Where: i represents the urban plot type; F i represents the frequency density of the i-th type of POI points in any urban area to the total number of POI points of this type; C i It represents the ratio of the frequency density of the i-th type of POI point in any urban plot to the frequency density of all types of POI points in the urban plot.

8. The multi-level building classification method based on multi-source spatial big data according to claim 6 is characterized in that: The expression of the spatial weight assignment algorithm is: Where: m represents the number of POI points of the i-th category within the radius range D with the center point of the building object as the center; w j Represents weight; N i Represents the total number of POI points of the i-th category in the POI dataset.

9. The multi-level building classification method based on multi-source spatial big data according to claim 1 is characterized in that: The step of acquiring the POI point data is: acquiring original data, and performing elimination and merging processing on the original data.

10. A multi-level building classification system based on multi-source spatial big data, characterized in that: include: A data acquisition module, used to acquire building vector data, traffic network data and POI point data, wherein the POI point data includes first POI point data and second POI point data; An urban plot division module, used for generating urban plot vector data according to the traffic road network data; A POI point classification module, configured to generate city plot classification data using the first POI point data according to the first mapping relationship; Based on the second mapping relationship, the building function classification data is generated using the second POI point data; the first mapping relationship is a mapping relationship between the first POI point data and the urban plot type; the second mapping relationship is a mapping relationship between the second POI point data and the building function type; The building classification module is used to assign building function types twice based on the building vector data and the urban plot vector data according to the urban plot classification data and the building function classification data.

Images