A grand canal hydraulic space type identification method based on multi-modal big data and fuzzy clustering algorithm

Abstract

The application discloses a grand canal water conservancy space type identification method based on a multi-modal big data and a fuzzy clustering algorithm, and belongs to the technical field of city space information processing. In view of the problems of unclear definition of the grand canal water conservancy space and low precision of a traditional investigation method, the application firstly collects building, land block, road and water system data through field investigation and an open map API to establish a multi-modal big database; then calculates the space intensity, density, maximum height and average height by taking the land block as a unit, and establishes the shortest distance relationship between the land block and the grand canal; finally, a fuzzy C means (FCM) clustering algorithm is used to cluster the space analysis unit to identify the water conservancy space type. The application realizes accurate quantitative identification and type division of the grand canal water conservancy space, can effectively process the space classification problem of a fuzzy boundary, and is suitable for fine management and protection of a linear heritage space.

Description

Technical Field

[0001] This invention relates to the field of urban spatial information processing technology, and in particular to a method for identifying the spatial types of the Grand Canal's hydraulic structures based on multimodal big data and fuzzy clustering algorithms. Background Technology

[0002] As a World Cultural Heritage site and a large-scale linear heritage site, the Grand Canal, due to its vast temporal and spatial span, advanced hydraulic engineering technology, and dynamic correlation with the development of settlements along its route, exhibits significant characteristics in its hydraulic space, including large scale, numerous elements, complex forms, and multifunctional applications. Existing methods for identifying and managing hydraulic spaces suffer from the following problems:

[0003] The definition and boundaries of the Grand Canal's hydraulic engineering space are unclear, and there is a lack of precise spatial identification methods.

[0004] Traditional survey and research methods have low precision and are difficult to accurately quantify large-scale complex spaces;

[0005] The existing data is not accurate enough to support detailed measurement and analysis of hydraulic spaces;

[0006] There is a lack of quantitative identification methods that can reveal the spatial morphology of hydraulic structures. Summary of the Invention

[0007] To address the problems existing in the prior art, this invention provides a method for accurate spatial identification of the Grand Canal's hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithms. By establishing a real and high-precision multimodal database, the FCM fuzzy clustering algorithm is applied to a certain spatial analysis unit to achieve the identification of the spatial types of the Grand Canal's hydraulic engineering structures and the accurate measurement of characteristic indicators.

[0008] The objective of this invention is achieved through the following technical solutions.

[0009] A method for identifying the spatial types of hydraulic engineering structures along the Grand Canal based on multimodal big data and fuzzy clustering algorithms includes the following steps:

[0010] Step 1: Establish a multimodal big data database for the Grand Canal, including collecting and processing multimodal data on buildings, land parcels, roads, and water systems, and importing it into a geographic information system to establish a hierarchical vector dataset;

[0011] Step 2: Create a spatial analysis unit layer based on plots of land, and summarize and calculate the spatial intensity, spatial density, maximum height, and average height of each plot;

[0012] Step 3: Establish the distance relationship between each plot and the Grand Canal, including creating centroids for spatial analysis units, extracting the Grand Canal channel line, calculating the shortest distance from the centroid to the channel, and associating the distance attribute with the spatial analysis unit;

[0013] Step 4: Use the fuzzy C-means clustering algorithm to cluster the spatial analysis units with distance attributes, and identify the spatial type of the Grand Canal hydraulic engineering based on the clustering results.

[0014] In step one, the collection of multimodal data includes: obtaining building and land use data of the urban section of the Grand Canal through field surveys; capturing vector geographic information through the Gaode Map Open Platform API and Python programming language; generating shapefile vector format through numerical calculations using human-computer interaction technology; and performing data cleaning, merging, and classification extraction.

[0015] In step one, the hierarchical vector dataset includes: building dataset, land parcel dataset, road dataset, and water system dataset; wherein, the building dataset includes building outline, number of building floors, building footprint, and total building area; the land parcel dataset includes land parcel outline and land parcel area; the road dataset includes road red line, road centerline, and road grade; and the water system dataset includes water system outline, river centerline, and water system name.

[0016] In step two,

[0017] The formula for calculating the spatial intensity of a land parcel is: FAR = / S, where Indicates the first The projected area of ​​each building. Indicates the first The number of floors in a building, S represents the area of ​​the land plot;

[0018] The formula for calculating the spatial density of a land parcel is: D = / S;

[0019] The formula for calculating the maximum height of a plot of land is: H = ;

[0020] The formula for calculating the average height of a plot of land is: H = .

[0021] Step three specifically includes:

[0022] Create a centroid for the spatial analysis unit layer to obtain the point layer "Layer_block_point";

[0023] Extract the Grand Canal waterway from the water system dataset based on the water system name to obtain a line layer;

[0024] Using the point layer as the source layer and the Grand Canal line layer as the target layer, a shortest line layer is created using vector analysis tools. This shortest line layer automatically includes distance attributes.

[0025] Using the field value concatenation tool, the distance attribute in the shortest line layer is associated with the point layer, resulting in a new layer "Layer_block_with_cannal_point" that contains the original spatial analysis unit attributes and the Grand Canal distance attribute.

[0026] Step four specifically includes:

[0027] In QGIS, call the doCluster function of the ClusterMap tool, select the fuzzy C-means clustering algorithm, input the spatial analysis unit point layer with distance attribute, and calculate to obtain a new point layer "Layer_FCM" containing the clustering attribute class;

[0028] Using the field value concatenation tool, the clustering attribute class is associated with the original spatial analysis unit layer to obtain a new layer "Layer_block_FCM";

[0029] The new layer “Layer_block_FCM” is symbolically categorized and displayed using the class field, and the clustering results are exported to complete the identification of the spatial type of the Grand Canal hydraulic engineering.

[0030] In step two, the spatial intensity of the land parcel is used to reflect the spatial development and construction situation on a unit of land; the spatial density of the land parcel is used to reflect the building coverage on a unit of land; the maximum height of the land parcel is used to reflect the vertical development of the land; and the average height of the land parcel is used to reflect the visual landscape pattern.

[0031] The Grand Canal's hydraulic spatial types include different categories based on the fuzzy C-means clustering algorithm, with each category corresponding to hydraulic spatial units that have similar spatial characteristics and distance characteristics.

[0032] The multimodal big data database is built on the QGIS geographic information system and adopts a unified data standard, coordinate system and display unit.

[0033] The spatial analysis unit is established on a plot-by-plot basis, and its attributes include plot spatial intensity, plot spatial density, plot maximum height, plot average height, and the shortest distance between the plot and the Grand Canal.

[0034] Compared with existing technologies, the advantages of this invention are: 1. This invention is the first to apply multimodal big data to linear heritage spatial identification: by comprehensively utilizing multimodal data of buildings, plots, roads, and water systems, a high-precision hydraulic spatial big data database is established.

[0035] 2. This invention innovatively introduces the shortest distance of the Grand Canal into the spatial analysis unit to identify the hydraulic space type, ensuring that the characteristics of the Grand Canal's hydraulic space are expressed.

[0036] 3. This invention innovatively uses the FCM (Fuzzy C-Means) fuzzy clustering algorithm to identify complex hydraulic spatial types: Fuzzy clustering can effectively handle the problem of spatial type division with fuzzy boundaries, and is more in line with the actual situation than traditional hard clustering. Attached Figure Description

[0037] Figure 1 This invention relates to the multimodal big data layer and its constituent elements;

[0038] Figure 2 This is a schematic diagram of the establishment of a spatial analysis unit in a certain urban area;

[0039] Figure 3 This is a schematic diagram showing the distance relationship between a plot of land in a certain urban area and the Grand Canal;

[0040] Figure 4 This is a schematic diagram showing the spatial type identification results and distribution of the Grand Canal in a certain urban area.

[0041] Figure 5 This is a flowchart of the present invention. Detailed Implementation

[0042] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0043] like Figure 5 As shown, a method for identifying the spatial types of the Grand Canal's hydraulic structures based on multimodal big data and fuzzy clustering algorithms includes the following steps:

[0044] Step 1: Establish a multimodal big data database for the Grand Canal

[0045] 1) Data Acquisition and Processing

[0046] Accurate building and land use data were obtained through on-site surveys of the urban section of the Grand Canal.

[0047] Vector geographic information is extracted using the Gaode Map Open Platform (API) and the Python programming language;

[0048] Using human-computer interaction technology, rapid numerical calculations are performed to generate shapefile vector format;

[0049] Perform data cleaning, merging, and classification extraction.

[0050] 2) Establishment of a multimodal database ( Figure 1 (As shown)

[0051] Import the categorized data into the QGIS geographic information system, establish a unified data standard, and form a unified data format, coordinate system, and display unit.

[0052] Establish a hierarchical vector dataset, including: building dataset (building outline, number of building floors, building footprint, total building area), land plot dataset (land plot outline, land plot area), road dataset (road boundary line, road center line, road grade), and water system dataset (water system outline, river center line, water system name).

[0053] Step 2: Establish spatial analysis units

[0054] Based on the multimodal big data of the Grand Canal, a spatial analysis unit layer "Layer_block" is established with land parcels as the unit, and the spatial intensity, spatial density, maximum height and average height of land parcels are calculated.

[0055] FAR (Facility Area Ratio) is the ratio of the total building area within a plot to the plot area, reflecting the spatial development and construction on a unit of land. The formula is: FAR = / S, where Indicates the first The projected area of ​​each building. Indicates the first The number of floors in a building, S represents the area of ​​the land plot;

[0056] Land space density is the ratio of the projected area of ​​buildings within a land parcel to the total land area. It reflects the degree of building coverage per unit of urban land. The formula is: D = / S, where Indicates the first The projected area of ​​a building, where S represents the plot area;

[0057] The maximum height of a plot of land is the peak building height within the plot, used to reflect the degree of vertical development of urban land. The calculation formula is: H = ,in Indicates the first The number of floors in each building;

[0058] The average height of a plot of land is the average building height within the block, used to reflect the visual landscape pattern. The calculation formula is: H = ,in Indicates the first The number of floors in a building.

[0059] Step 3: Establish the distance relationship between the plot and the Grand Canal

[0060] Create a centroid for the spatial analysis unit established in step two to obtain a new point layer "Layer_block_point" with the same attributes as the original spatial analysis unit.

[0061] Extract the Grand Canal waterways from the water system dataset based on their names to create a new line layer;

[0062] Using the spatial analysis unit centroid point layer as the source layer and the Grand Canal line layer as the target layer, create a shortest line layer using the "Vector Analysis - Shortest Line Between Features" tool. This shortest line layer will automatically include the "distance" attribute.

[0063] Using the "Vector General - Connect Attributes by Field Value" tool, with the centroid point of the spatial analysis unit as the input layer, "OBJECTID" as the table field, and the shortest line as the input layer 2, with "OBJECTID" as the table field 2, select "distance" for the field of layer 2 to be copied, execute the algorithm and save it as a new layer "Layer_block_with_cannal_point". The new layer will contain the original spatial analysis unit attributes and the Grand Canal distance attributes, thus completing the establishment of the distance relationship between the plot and the Grand Canal.

[0064] Step 4: Identify hydraulic spatial types using the FCM fuzzy clustering algorithm.

[0065] In QGIS, use the "ClusterMap-doCluster" tool with the following parameter settings: select "Layer_block_with_cannal_point" for point layer, select "Fuzzy C-Means" for cluster algorithm, and leave other parameters as default to obtain a new point layer "Layer_FCM".

[0066] Using the "Vector General - Connect Attributes by Field Value" tool, with "Layer_block" as the input layer and "OBJECTID" as the table field, and "Layer_FCM" as the input layer 2 and "OBJECTID" as the table field 2, select "class" for the field of layer 2 to be copied, execute the algorithm and save it as a new layer "Layer_block_FCM". The new layer will contain the original spatial analysis unit attributes and FCM clustering attributes.

[0067] The “Layer_block_FCM” is symbolically displayed. Select “Classification”, use the “class” field as the value, display and export the clustering results to complete the identification of the spatial type of the Grand Canal hydraulic engineering.

[0068] Example

[0069] 1) By conducting on-site surveys and using the Gaode Map API, vector data of buildings, plots, roads, and water systems in a certain urban area were obtained, and a multimodal big data database containing 143,590 buildings and 3,214 plots was established.

[0070] 2) Create a spatial analysis unit layer "Layer_block" for each plot, and summarize and calculate the plot spatial intensity, plot spatial density, plot maximum height, and plot average height. Figure 2 (As shown). Among them, the maximum spatial intensity of the plot is 10.68, and the average is 0.18; the maximum spatial density of the plot is 0.98, and the average is 0.04; the maximum height of the plot is 69 stories, and the average is 1.51 stories; the maximum average height of the plot is 30.3 stories, and the average is 0.84 stories.

[0071] 3) Create a centroid layer “Layer_block_point” using the spatial analysis unit established above, with attributes identical to those of the original spatial analysis unit; extract the Grand Canal channel line layer from the water system dataset based on its name; create a shortest line layer using “Layer_block_point” as the source layer and the Grand Canal channel line layer as the target layer; connect attributes by field value, using “Layer_block_point” as the input layer and “OBJECTID” as the table field, and the shortest line layer as input layer 2, using “OBJECTID” as table field 2, selecting “distance” as the field to be copied, and save it as “Layer_block_with_cannal_point”. This layer contains the attributes of the original spatial analysis unit and the distance attribute to the Grand Canal, completing the establishment of the distance relationship between the plot and the Grand Canal. Figure 3 (As shown).

[0072] 4) Using "ClusterMap", input the point layer "Layer_block_with_cannal_point", select "Fuzzy C-Means" as the clustering method, and leave other parameters as default to obtain "Layer_FCM"; connect the attributes by field value, using "Layer_block" as the input layer and "OBJECTID" as the table field, and "Layer_FCM" as the second input layer and "OBJECTID" as the second table field. Select "class" as the field to be copied, and save it as "Layer_block_FCM". This layer will contain the original spatial analysis unit attributes and FCM clustering attributes.

[0073] 5) To display the "Layer_block_FCM" symbolically, select "Classification," use the "class" field as the value, and display and export the clustering results. Figure 4 As shown in the figure, the spatial type of the Grand Canal in a certain urban area was identified.

Claims

1. A method for identifying the spatial types of hydraulic engineering structures along the Grand Canal based on multimodal big data and fuzzy clustering algorithms, characterized in that: Includes the following steps: Step 1: Establish a multimodal big data database for the Grand Canal, including collecting and processing multimodal data on buildings, land parcels, roads, and water systems, and importing it into a geographic information system to establish a hierarchical vector dataset; Step 2: Create a spatial analysis unit layer based on plots of land, and summarize and calculate the spatial intensity, spatial density, maximum height, and average height of each plot; Step 3: Establish the distance relationship between each plot and the Grand Canal, including creating centroids for spatial analysis units, extracting the Grand Canal channel line, calculating the shortest distance from the centroid to the channel, and associating the distance attribute with the spatial analysis unit; Step 4: Use the fuzzy C-means clustering algorithm to cluster the spatial analysis units with distance attributes, and identify the spatial type of the Grand Canal hydraulic engineering based on the clustering results.

2. The method for identifying the spatial types of the Grand Canal's hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, In step one, the collection of multimodal data includes: obtaining building and land use data of the urban section of the Grand Canal through field surveys; capturing vector geographic information through the Gaode Map Open Platform API and Python programming language; generating shapefile vector format through numerical calculations using human-computer interaction technology; and performing data cleaning, merging, and classification extraction.

3. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, In step one, the hierarchical vector dataset includes: building dataset, land parcel dataset, road dataset, and water system dataset; wherein, the building dataset includes building outline, number of building floors, building footprint, and total building area; the land parcel dataset includes land parcel outline and land parcel area; the road dataset includes road red line, road centerline, and road grade; and the water system dataset includes water system outline, river centerline, and water system name.

4. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, In step two The formula for calculating the spatial intensity of a land parcel is: FAR = / S, where Indicates the first The projected area of ​​each building. Indicates the first The number of floors in a building, S represents the area of ​​the land plot; The formula for calculating the spatial density of a land parcel is: D = / S; The formula for calculating the maximum height of a plot of land is: H = ; The formula for calculating the average height of a plot of land is: H = .

5. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, Step three specifically includes: Create a centroid for the spatial analysis unit layer to obtain the point layer "Layer_block_point"; Extract the Grand Canal waterway from the water system dataset based on the water system name to obtain a line layer; Using the point layer as the source layer and the Grand Canal line layer as the target layer, a shortest line layer is created using vector analysis tools. This shortest line layer automatically includes distance attributes. Using the field value concatenation tool, the distance attribute in the shortest line layer is associated with the point layer, resulting in a new layer "Layer_block_with_cannal_point" that contains the original spatial analysis unit attributes and the Grand Canal distance attribute.

6. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, Step four specifically includes: In QGIS, call the doCluster function of the ClusterMap tool, select the fuzzy C-means clustering algorithm, input the spatial analysis unit point layer with distance attribute, and calculate to obtain a new point layer "Layer_FCM" containing the clustering attribute class; Using the field value concatenation tool, the clustering attribute "class" is associated with the original spatial analysis unit layer to obtain a new layer "Layer_block_FCM"; The new layer "Layer_block_FCM" is symbolically categorized and displayed using the class field, and the clustering results are exported to complete the identification of the spatial type of the Grand Canal hydraulic engineering.

7. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, In step two, the spatial intensity of the land parcel is used to reflect the spatial development and construction situation on a unit of land; the spatial density of the land parcel is used to reflect the building coverage on a unit of land; the maximum height of the land parcel is used to reflect the vertical development of the land; and the average height of the land parcel is used to reflect the visual landscape pattern.

8. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, The Grand Canal's hydraulic spatial types include different categories based on the fuzzy C-means clustering algorithm, with each category corresponding to hydraulic spatial units that have similar spatial characteristics and distance characteristics.

9. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, The multimodal big data database is built on the QGIS geographic information system and adopts a unified data standard, coordinate system and display unit.

10. The method for identifying the spatial types of the Grand Canal hydraulic engineering structures based on multimodal big data and fuzzy clustering algorithm according to claim 1, characterized in that, The spatial analysis unit is established on a plot-by-plot basis, and its attributes include plot spatial intensity, plot spatial density, plot maximum height, plot average height, and the shortest distance between the plot and the Grand Canal.

Images