Method and system for identifying buildings based on building attributes and geospatial information

Abstract

The application relates to the field of house building industry, and discloses a method and system for identifying buildings based on house building attributes and geographic space information, which comprises the following steps: determining house building attributes according to house building standards; classifying the house building attributes into special information, important information, geographic space information and basic information according to the house building attribute recognition degree; setting corresponding matching values and constraint conditions for the house building attributes of different types according to the house building types; calculating the coincidence degree between a to-be-identified house building graph spot and the house building graph spots in the system, multiplying the coincidence degree by the matching values to obtain a modified house building space geographic position matching score value; further calculating the matching total score between the to-be-identified house building and the house buildings in the system, and identifying the same house, similar house or new house. According to the application, the actual geographic space position of the house is analyzed based on the graph spot position provided by the house, and more accurate graph spot identification is realized.

Description

Technical Field

[0001] This invention relates to the field of building construction, specifically to a method and system for identifying buildings based on building attributes and geospatial information. Background Technology

[0002] Currently, standard building codes in the construction industry cannot uniquely and accurately identify buildings. At the same geographical location, there may be both underground and above-ground structures. Furthermore, the identity of buildings at the same geographical location is not entirely the same across different systems; there will always be some discrepancies in building maps between different systems. Therefore, it is necessary to combine other building attributes to accurately identify the building's identity. Summary of the Invention

[0003] To address the problems existing in the prior art, this invention proposes a method for identifying buildings based on building attributes and geospatial information. Based on the characteristics of the construction industry, the method analyzes the location of the map patches provided by the building and the actual geospatial location of the building to achieve more accurate map patch identification, ultimately enabling the identification of a unique identity for the same building.

[0004] On the other hand, the present invention also provides a system for identifying buildings based on building attributes and geospatial information.

[0005] The method for identifying buildings based on building attributes and geospatial information in this embodiment of the invention includes the following steps:

[0006] The building attributes are determined according to building construction standards;

[0007] Based on the identifiability of building attributes, building attributes are classified. The types of building attributes after classification include special information, important information, geospatial information, and basic information. Among them, special information is attribute information that can uniquely identify the identity of a building; important information is attribute information that can locate a specific building; geospatial information is attribute information that identifies the building in a specific geographical location; and basic information is the building's construction information. The geospatial information includes the building's spatial geographical location.

[0008] Based on the building type, set corresponding matching values ​​and constraints for different types of building attributes;

[0009] Calculate the overlap between the building patch to be identified and the building patches in the system, and multiply the calculated overlap by the matching value set for the building's spatial geographic location to obtain the corrected building spatial geographic location matching score.

[0010] Based on the set matching values ​​of various building attributes and the corrected matching score of building spatial location, the total matching score between the building to be identified and the buildings in the system is calculated, and it is identified as the same building, a similar building, or a new building.

[0011] Preferably, the matching value is defined based on the similarity of the building attributes; the more obvious the attribute features, the higher the matching value is set.

[0012] The constraints on building attributes include M, C, and O; constraint M indicates that the corresponding building attribute is mandatory; constraint C indicates that the corresponding building attribute is mandatory when the conditions are met; constraint O indicates that the corresponding building attribute is optional.

[0013] Preferably, the degree of overlap is measured by calculating the Jaccard similarity coefficient between the building patch to be identified and the building patches in the system; when both the building patch to be identified and the building patches in the system are empty, the Jaccard similarity coefficient is defined as 1.

[0014] Preferably, the step of correcting the geographical location matching score of the building's spatial location includes:

[0015] Obtain the building patch data to be identified and the building patch data in the system, and create a new area field;

[0016] Convert the geographic coordinate system of the patch data into a projected coordinate system;

[0017] Calculate the area of ​​the polygon and update the calculated area value to the area field;

[0018] Calculate the intersection between the building feature to be identified and the building features in the system, and calculate the intersection area;

[0019] Calculate the Jaccard similarity coefficient between the building patch to be identified and the building patches in the system;

[0020] The final building spatial location matching score is calculated based on the Jaccard similarity coefficient.

[0021] Preferably, the formula for calculating the total matching score between the building to be identified and the buildings in the system is as follows:

[0022] Total(H i H j ) = Sim(H i H j )+Sim(S i ,S j )

[0023] Sim(Hi H j Sim(S) matches the accumulated value of the attribute item. i ,S j () represents the corrected score for matching the geographical location of the building's spatial location.

[0024] The system for identifying buildings based on building attributes and geospatial information provided in this embodiment of the invention includes the following modules:

[0025] The building attribute acquisition module determines the building attributes based on building standards.

[0026] The building attribute classification module categorizes building attributes based on their identifiability. The categorized building attribute types include special information, important information, geospatial information, and basic information. Special information uniquely identifies a building; important information pinpoints a specific building; geospatial information identifies the building's location; and basic information is the building's construction information. The geospatial information includes the building's geographical location.

[0027] The matching rule definition module sets corresponding matching values ​​and constraints for different types of building attributes based on the building type.

[0028] The spatial geographic location matching value correction module calculates the overlap between the building patch to be identified and the building patches in the system, and multiplies the calculated overlap by the matching value set for the building's spatial geographic location to obtain the corrected building spatial geographic location matching score.

[0029] The identification module calculates the total matching score between the building to be identified and the buildings in the system based on the set matching values ​​of various building attributes and the corrected matching score of the building's spatial location, and identifies it as the same building, a similar building, or a new building.

[0030] Compared with the prior art, the technical effects achieved by the present invention include:

[0031] 1. This invention achieves building identification and unique coding through the established similarity recognition model and defined related recognition algorithms. Ultimately, the building code is unified as a unified identity identifier for buildings, which is unique in time and space and runs through the entire life cycle of building design, procurement, construction, transaction, operation and maintenance and demolition.

[0032] 2. This invention assigns a unique and permanent identification code to all buildings, effectively preventing the same building from having different building codes in different systems, which would prevent data from being interconnected between different systems.

[0033] 3. This invention can effectively achieve the advantage of unique coding of buildings across different systems, providing basic core data support for digital building lifecycle management and corresponding information system applications. Attached Figure Description

[0034] Figure 1 This is a flowchart of the building identification method in an embodiment of the present invention;

[0035] Figure 2 This is a schematic diagram of the coding rules for building codes in an embodiment of the present invention. Detailed Implementation

[0036] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments, but the implementation of the present invention is not limited thereto.

[0037] Example 1

[0038] This embodiment provides a method for identifying buildings based on building attributes and geospatial information. Before assigning building codes, the system accurately identifies different buildings. Based on the basic attribute information and geospatial information of the buildings, the system constructs a unique building model library. In the existing building code library, some predefined rules and algorithms are automatically executed to accurately identify different buildings, and a unique building code is assigned to the identified buildings or an existing building code is issued.

[0039] Specifically, such as Figure 1 As shown, the method in this embodiment includes the following steps:

[0040] Step S1: Determine the building attributes; According to building standards, building attributes mainly include project code, building name, standard address, detailed address, construction project planning permit number or rural construction planning permit number, construction project construction permit number, land parcel code, real estate unit code, structure type, temporary building, year of completion and acceptance, number of floors above ground, number of floors below ground, building height, ground parking spaces, underground parking spaces, base area, total building area, main use, protected building, demolition, special instructions, and building spatial location (spatial map information with coordinate system), etc.

[0041] Step S2: Classify building attributes based on their identifiability. The types of building attributes after classification include special information, important information, geospatial information, and basic information. Special information refers to attribute information that can uniquely identify a building, important information refers to attribute information that can locate a specific building, geospatial information refers to attribute information that identifies the building in a specific geographical location, and basic information refers to the building's construction information, as shown in Table 1 below.

[0042] Table 1 Classification of building attributes based on recognizability

[0043]

[0044] Step S3: Based on the building type, set corresponding matching values ​​and constraints for the building attributes of different types of buildings.

[0045] As shown in Tables 2-5 below, the constraints on the building attributes include M, C, and O. Constraint M indicates that the corresponding building attribute is mandatory, i.e., it must be present; constraint C indicates that the corresponding building attribute is mandatory when the conditions are met, i.e., it should be present when the actual situation is met; constraint O indicates that the corresponding building attribute is optional, i.e., it can be determined by the user whether it is needed.

[0046] In this embodiment, the matching value is set manually based on the similarity of the building attributes. The more obvious the attribute features, the higher the defined and set matching value.

[0047] The matching values, or matching scores, for building attribute types "Special Information" are denoted as set A = {1, 2, 3, 4, 5}, as shown in Table 2. In this embodiment, the constraint condition for the building attribute "Special Information" is always C; any corresponding attribute information must be filled in.

[0048] Table 2 Matching values ​​and constraints for special information

[0049]

[0050] The matching values ​​for the building attribute type "Important Information" are denoted as set B = {6, 7, 8}, as shown in Table 3. Within the Important Information, the constraint for the standard address is M, the constraint for the building name is C, and the constraint for the detailed address is O.

[0051] Table 3 Matching values ​​and constraints for important information

[0052]

[0053] The matching values ​​for the building attribute type "geospatial information" are denoted as set C = {9}, as shown in Table 4. This type of attribute includes the building's spatial geographic location, and its constraint is M.

[0054] Table 4 Matching values ​​and constraints of geospatial information

[0055]

[0056] In this embodiment, the basic information mainly includes the construction information and related descriptions of the building itself. The matching values ​​of the building attribute type "basic information" are denoted as set D = {10,11,12,13,14,15,16,17,18,19,20,21,22,23}, as shown in Table 5. The constraints for attributes such as base area, total building area, structural type, building height, protected building status, and demolition status are M. For most buildings, the protected building and demolition attributes are both "no." In this embodiment, these two attributes are only used to mark special buildings, and therefore their matching values ​​are relatively low.

[0057] Table 5 Matching values ​​and constraints for basic information

[0058]

[0059] In this embodiment, the matching score of the building name in the important information in Table 3 is not fixed, but is set differently according to the different uses of the building, as shown in Table 6 below.

[0060] Table 6. Matching scores of building names for different uses

[0061]

[0062]

[0063] Most rural houses do not have official house names, so their matching scores are relatively low. Urban houses have higher matching scores than rural houses because they are mostly residential communities and large-scale residential buildings, which usually have corresponding community names, making it easier to identify and locate the building based on the building name. Commercial, industrial, and medical buildings usually have clear building names, making it easier for the general public to identify and locate the building, so their matching scores are relatively high.

[0064] The matching score for the year of completion and acceptance in Table 5 is not fixed, but is set to different levels and defined with different matching score values ​​according to different years of completion and acceptance; see Table 7 below for details:

[0065] Table 7 Matching scores for different completion and acceptance years

[0066]

[0067] For houses that were generally completed and inspected within the last 15 years, the identification of different houses based on the completion and inspection year is not obvious, so the matching score is set to 5; however, as the years increase, the identification gradually becomes stronger, and the matching score set for them also gradually increases.

[0068] Similarly, the matching score for building height in Table 5 is not fixed, but is set to different levels and defined with different matching scores based on the building height itself; see Table 8 below for details:

[0069] Table 8 Matching scores for different building heights

[0070]

[0071] Buildings less than or equal to 10m are generally 2-3 stories high, and the height difference is not very noticeable, so the matching score is set to 5. Buildings greater than 10m but less than 20m are generally 6-8 stories high, and are more easily identifiable than buildings less than 10m, so the matching score is set to 10. Similarly, buildings greater than or equal to 20m are more easily identifiable, so the matching score is set to 15.

[0072] Step S4: Calculate the overlap between the building patch to be identified and the building patches in the system. Multiply the calculated overlap by the matching value set for the building's spatial geographic location to obtain the corrected building spatial geographic location matching score, thereby improving the matching accuracy of the building's spatial geographic location.

[0073] In this embodiment, a patch refers to a vector polygon of the base map of a building, and a patch is also called a layer; the matching score is highest for the building attribute of the building's spatial location, so the matching accuracy of the geographic location is particularly important.

[0074] Based on this, the specific geospatial location matching algorithm is as follows:

[0075] The degree of overlap between two building patches is calculated using tools or algorithms to evaluate the geospatial matching of the two buildings. The degree of overlap is measured by the Jaccard similarity coefficient between the building patch to be identified and other building patches in the system. The Jaccard similarity coefficient is used to compare the similarity and differences between a finite set of samples; a higher Jaccard similarity coefficient indicates a higher degree of sample similarity.

[0076] Given a census survey patch A (i.e., a building patch in the system) and a coded building patch B (i.e., a building patch to be identified), the Jaccard similarity coefficient is defined as the ratio between the intersection of patch A and patch B and the union of patch A and patch B, and the formula is as follows:

[0077]

[0078] When both polygons A and B are empty, the Jaccard similarity coefficient J(A,B) is defined as 1.

[0079] In house similarity scoring, the Jaccard similarity coefficient can be calculated by taking the intersection and union of the census survey map patches and the coded house map patches. The closer the similarity coefficient is to 1, the higher the house similarity and the higher the score.

[0080] PostGIS is an extension of the object-relational database system PostgreSQL. PostGIS provides the following spatial information services: spatial objects, spatial indexes, spatial operation functions, and spatial operators. Furthermore, PostGIS conforms to the OpenGIS specification. Using PostGIS, one can directly manage and edit census survey patches and coded building patches, and perform geospatial information analysis, data sharing, and display.

[0081] In other words, this embodiment first calculates the actual areas of the two map features—the survey map feature and the coded house map feature—then calculates the intersection area of ​​the two map features, and finally calculates the Jaccard coefficient value using the Jaccard coefficient definition formula. The closer the coefficient value is to 1, the higher the similarity. In this embodiment, this step specifically includes:

[0082] (1) Obtain census survey map data and coded house map data, and create a new area field.

[0083] (2) Since the coordinate system of the patch data is srid = 4490, which is a geographic coordinate system, it is necessary to convert the geographic coordinate system of the patch data into a projected coordinate system if the length and area are to be calculated. In this embodiment, it is converted to srid = 4527.

[0084] (3) Use the ST_Area function to calculate the area of ​​the polygon and update the calculated area value to the area field. The SQL is as follows:

[0085] update risk_census_house_137100SET area=

[0086] ST_Area(st_transform(st_setsrid(geom,4490),4527));

[0087] update house_code_137100SET area=ST_Area(st_transform(st_setsrid(geom,4490),4527)).

[0088] (4) Use the ST_Intersection function to calculate the intersection of two polygons and calculate the intersection area. The results are shown in Table 9. The key SQL is as follows:

[0089] st_area(st_transform(st_setsrid(ST_Intersection(a1.geom1,a1.geom2),4490),4527))

[0090] Table 9 shows the areas of two polygons and their intersection area.

[0091]

[0092] (5) Calculate the Jaccard similarity coefficient between the two patches. The results are shown in Table 10 below:

[0093] Table 10 Jaccard similarity coefficients of two map features

[0094]

[0095] (6) The final building spatial location matching score was calculated based on the Jaccard similarity coefficient, and the results are as follows:

[0096] If Jaccard = 1, it means that the two patches completely overlap, and the score is 60 * 100% = 60 points, where 60 is the matching value set for the spatial geographical location of the building in step S3, and 100% is the Jaccard similarity coefficient.

[0097] When Jaccard ≥ 0.9, the score is 60 * 90% = 54 points, where 90% is the Jaccard similarity coefficient;

[0098] When 0.9 > Jaccard ≥ 0.8, the score is 60 * 80% = 48 points, where 80% is the Jaccard similarity coefficient;

[0099] When 0.8 > Jaccard ≥ 0.7, the score is 60 * 70% = 42 points, where 70% is the Jaccard similarity coefficient;

[0100] When Jaccard < 0.7, it means that the two patches have little overlap, and the score is defined as 0.

[0101] Step S5: Based on the matching values ​​of various building attributes set in Step S3 and the matching score of building spatial location corrected in Step S4, calculate the total matching score between the building to be identified and the buildings in the system, and identify it as the same building, a similar building, or a new building.

[0102] In this embodiment, when calculating the total matching score, spatial attributes (i.e., spatial geographical location) are calculated separately, while the matching values ​​of other building attributes besides spatial attributes are dynamically accumulated to obtain the cumulative matching value of the attribute items. Based on extensive research and verification of existing building data in the system, the following calculation formula for dynamic accumulation is summarized:

[0103]

[0104] Among them, Sim(H i H j The sum of the matching values ​​for each building attribute is the cumulative value obtained after dynamically accumulating the matching values ​​for each building attribute; a k This represents the matching value corresponding to the k-th building attribute item, where there are several items 'a'. k The value is dynamically calculated, obtained by matching the specific attribute value provided by the user against the matching rules defined in step S3; Sim(H ik H jk Let be the similarity between the building to be identified and the k-th attribute item of the buildings in the system. In this embodiment, the similarity of attribute items is defined as follows: if the attribute items are the same, the similarity is defined as 1; if the attribute items are different, the similarity is defined as 0. That is to say, for the comparison of attributes other than spatial attributes, there is only the case of being the same or different. If they are the same, the matching value of that attribute item is added to the total matching score; if they are different, the matching value of that attribute item is not considered.

[0105] Given that the geospatial information of buildings may have certain deviations or errors in different tools, this embodiment considers the similarity of spatial attributes separately. The similarity (i.e., overlap) of spatial attributes is calculated using the Jaccard similarity coefficient. The method for calculating the spatial geographic location matching score is as follows:

[0106] Sim(S i ,S j ) = 60 * Sim(S i1 ,S j1 )

[0107] Among them, Sim(S i ,S jSim(S) represents the matching score of the building's spatial attributes, i.e., the matching score of the building's geographical location after correction in step S4; i1 ,S j1 The overlap between the building to be identified and the buildings in the system is the degree of geographical spatial overlap. In this embodiment, the effective numerical range of overlap is defined as 70% to 100%. When the overlap is less than 70%, it is directly set to 0, that is, it is not considered when calculating the total matching score. For the specific calculation method, please refer to step S4.

[0108] Therefore, the formula for calculating the total matching score between the building to be identified and the buildings in the system is as follows:

[0109] Total(H i H j ) = Sim(H i H j )+Sim(S i ,S j )

[0110] This embodiment is based on extensive verification of existing building data in the system, and finally arrives at a more reasonable judgment of similar houses, as shown in Table 11 below.

[0111] Table 11: Total Score Ranking for Matching Similar Houses

[0112]

[0113] 1) Conditions under which the system identifies the property as the same house

[0114] To ensure the accuracy of house similarity identification, in addition to a total matching score of 130 or higher, and to prevent inaccurate house similarity identification due to a single building attribute type, the following requirements apply to the matching item attributes (referred to as conditions Con_attr):

[0115] The attribute set A1 of special information must belong to A (A1∈A), and the number of attributes in set A1 must be greater than or equal to 1;

[0116] The attribute set B1 of important information must belong to B (B1∈B), and the number of attributes in set B1 must be greater than or equal to 1;

[0117] The set of attributes C1 of geospatial information is a necessary item, therefore C1 = C;

[0118] The attribute set D1 of basic information must belong to D (D1∈D), and the number of attributes in set D1 must be greater than or equal to 3.

[0119] 2) Conditions under which the system identifies houses as similar

[0120] If the total matching score reaches 130 or above, but does not meet the Con_attr condition, the system identifies it as a similar house, and manual review is required to further confirm whether it is the same house.

[0121] If the total matching score is 80 or higher but less than 130, the system identifies it as a similar house, and manual review is required to further confirm whether it is the same house.

[0122] 3) The system identifies it as a new house.

[0123] If the total matching score is less than 80, the system will automatically identify it as a different house.

[0124] Step S6: When a property is identified as a new property or confirmed as a new property through manual review, a unique code is assigned to the new property according to the coding rules.

[0125] This embodiment defines the coding rules for building codes according to the building code standard as follows: Figure 2 As shown.

[0126] The building code should consist of 18 digits. The first 6 digits are the county-level administrative division code, the 7th digit is the year and month classification code, the 8th to 13th digits are the permitting month and month or the completion month and month code, and the 14th to 18th digits are the sequence code. The sequence code is 5 digits long and ranges from 00001 to 99999. It is automatically generated by a computer program. It should be ensured that the building code is not repeated, and it is advisable to start coding from 00001 every year.

[0127] Example 2

[0128] Based on the same inventive concept as Embodiment 1, this embodiment provides a system for identifying buildings based on building attributes and geospatial information, including the following modules:

[0129] The building attribute acquisition module determines the building attributes based on building standards.

[0130] The building attribute classification module categorizes building attributes based on their identifiability. The categorized building attribute types include special information, important information, geospatial information, and basic information. Special information uniquely identifies a building; important information pinpoints a specific building; geospatial information identifies the building's location; and basic information is the building's construction information. The geospatial information includes the building's geographical location.

[0131] The matching rule definition module sets corresponding matching values ​​and constraints for different types of building attributes based on the building type.

[0132] The spatial geographic location matching value correction module calculates the overlap between the building patch to be identified and the building patches in the system, and multiplies the calculated overlap by the matching value set for the building's spatial geographic location to obtain the corrected building spatial geographic location matching score.

[0133] The identification module calculates the total matching score between the building to be identified and the buildings in the system based on the set matching values ​​of various building attributes and the corrected matching score of the building's spatial location, and identifies it as the same building, a similar building, or a new building.

[0134] In this embodiment, the step of correcting the spatial geographic location matching score of a building in the spatial geographic location matching value correction module includes:

[0135] Obtain the building patch data to be identified and the building patch data in the system, and create a new area field;

[0136] Convert the geographic coordinate system of the patch data into a projected coordinate system;

[0137] Calculate the area of ​​the polygon and update the calculated area value to the area field;

[0138] Calculate the intersection between the building feature to be identified and the building features in the system, and calculate the intersection area;

[0139] Calculate the Jaccard similarity coefficient between the building patch to be identified and the building patches in the system;

[0140] The final building spatial location matching score is calculated based on the Jaccard similarity coefficient.

[0141] The formula for calculating the total matching score between the building to be identified and the buildings in the system is as follows:

[0142] Total(H i H j ) = Sim(H i H j )+Sim(S i ,S j )

[0143] Sim(H i H j Sim(S) matches the accumulated value of the attribute item. i ,S j () represents the corrected score for matching the geographical location of the building's spatial location;

[0144] The attribute item matching cumulative value is obtained by dynamically accumulating the matching values ​​of all building attributes except for the building's spatial geographical location attribute.

[0145] The modules described above in this embodiment are used to implement the corresponding steps in Embodiment 1. For detailed implementation process of each module, please refer to the description of Embodiment 1.

[0146] The above embodiments of the present invention are merely examples for clearly illustrating the present invention and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively describe all possible implementations. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for identifying buildings based on building attributes and geospatial information, characterized in that, Includes the following steps: The building attributes are determined according to building construction standards; Based on the identifiability of building attributes, building attributes are classified. The types of building attributes after classification include special information, important information, geospatial information, and basic information. Among them, special information is attribute information that can uniquely identify the identity of a building; important information is attribute information that can locate a specific building; geospatial information is attribute information that identifies the building in a specific geographical location; and basic information is the building's construction information. The geospatial information includes the building's spatial geographical location. Based on the building type, set corresponding matching values ​​and constraints for different types of building attributes; Calculate the overlap between the building patch to be identified and the building patches in the system, and multiply the calculated overlap by the matching value set for the building's spatial geographic location to obtain the corrected building spatial geographic location matching score. Based on the set matching values ​​of various building attributes and the corrected matching score of building spatial location, the total matching score between the building to be identified and the buildings in the system is calculated, and it is identified as the same building, a similar building, or a new building.

2. The method for identifying buildings according to claim 1, characterized in that, The matching value is defined based on the similarity of the building attributes; the more obvious the attribute characteristics, the higher the matching value is set. The constraints on building attributes include M, C, and O; Constraint M indicates that the corresponding building attribute is mandatory; constraint C indicates that the corresponding building attribute is mandatory when the conditions are met; constraint O indicates that the corresponding building attribute is optional.

3. The method for identifying buildings according to claim 1, characterized in that, The degree of overlap is measured by calculating the Jaccard similarity coefficient between the building patch to be identified and the building patches in the system; when both the building patch to be identified and the building patches in the system are empty, the Jaccard similarity coefficient is defined as 1.

4. The method for identifying buildings according to claim 1, characterized in that, The steps to correct the building's spatial location matching score include: Obtain the building patch data to be identified and the building patch data in the system, and create a new area field; Convert the geographic coordinate system of the patch data into a projected coordinate system; Calculate the area of ​​the polygon and update the calculated area value to the area field; Calculate the intersection between the building feature to be identified and the building features in the system, and calculate the intersection area; Calculate the Jaccard similarity coefficient between the building patch to be identified and the building patches in the system; The final building spatial location matching score is calculated based on the Jaccard similarity coefficient.

5. The method for identifying buildings according to claim 4, characterized in that, When the Jaccard similarity coefficient is less than 0.7, the spatial location matching score of the building is defined as 0.

6. The method for identifying buildings according to any one of claims 1-5, characterized in that, The formula for calculating the total matching score between the building to be identified and the buildings in the system is as follows: Total(H i ,H j )=Sim(H i ,H j )+Sim(S i ,S j ) Sim(H i H j Sim(S) matches the accumulated value of the attribute item. i ,S j () represents the corrected score for matching the geographical location of the building's spatial location.

7. The method for identifying buildings according to claim 6, characterized in that, When calculating the total matching score, the matching values ​​of all building attributes except for the building's spatial location attribute are dynamically accumulated to obtain the cumulative matching value for each attribute. The formula for this dynamic accumulation is as follows: Among them, Sim(H i H j ) matches the accumulated value of the attribute item; a k This represents the matching value corresponding to the k-th building attribute item; Sim(H ik H jk ) represents the similarity between the building to be identified and the k-th attribute item of the buildings in the system.

8. The method for identifying buildings according to claim 7, characterized in that, The similarity of attribute items is defined as follows: if the attribute items are the same, the similarity is defined as 1; if the attribute items are different, the similarity is defined as 0.

9. A system for identifying buildings based on building attributes and geospatial information, characterized in that, Includes the following modules: The building attribute acquisition module determines the building attributes based on building standards. The building attribute classification module categorizes building attributes based on their identifiability. The categorized building attribute types include special information, important information, geospatial information, and basic information. Special information uniquely identifies a building; important information pinpoints a specific building; geospatial information identifies the building's location; and basic information is the building's construction information. The geospatial information includes the building's geographical location. The matching rule definition module sets corresponding matching values ​​and constraints for different types of building attributes based on the building type. The spatial geographic location matching value correction module calculates the overlap between the building patch to be identified and the building patches in the system, and multiplies the calculated overlap by the matching value set for the building's spatial geographic location to obtain the corrected building spatial geographic location matching score. The identification module calculates the total matching score between the building to be identified and the buildings in the system based on the set matching values ​​of various building attributes and the corrected matching score of the building's spatial location, and identifies it as the same building, a similar building, or a new building.

10. The system for identifying buildings according to claim 9, characterized in that, In the spatial location matching value correction module, the steps for correcting the spatial location matching score of buildings include: Obtain the building patch data to be identified and the building patch data in the system, and create a new area field; Convert the geographic coordinate system of the patch data into a projected coordinate system; Calculate the area of ​​the polygon and update the calculated area value to the area field; Calculate the intersection between the building feature to be identified and the building features in the system, and calculate the intersection area; Calculate the Jaccard similarity coefficient between the building patch to be identified and the building patches in the system; The final building spatial location matching score is calculated based on the Jaccard similarity coefficient. The formula for calculating the total matching score between the building to be identified and the buildings in the system is as follows: Total(H i ,H j )=Sim(H i ,H j )+Sim(S i ,S j ) Sim(H i H j Sim(S) matches the accumulated value of the attribute item. i ,S j () represents the corrected score for matching the geographical location of the building's spatial location; The attribute item matching cumulative value is obtained by dynamically accumulating the matching values ​​of all building attributes except for the building's spatial geographical location attribute.

Images