Massive housing risk hidden danger visual evaluation method and system

Abstract

The application belongs to the field of house data processing, and is a mass house risk hidden danger visualization evaluation method and system, which comprises the following steps: performing building comprehensive risk evaluation on a single house to calculate a risk hidden danger coefficient of the single house; performing spatial aggregation processing of different degrees on mass houses from multiple administrative region levels to obtain a risk hidden danger distribution situation of the aggregated point houses after aggregation; converting discrete vector points into continuous grid network data for display through kernel density analysis; and performing result analysis and visual rendering display according to the heat space distribution characteristic grid network data. The application calculates heat distribution based on the density weight of the risk hidden danger index, enhances the display effect of the geographical space distribution characteristics of the house risk hidden danger index, and solves the technical problems of lag, full memory occupation and the like during the rendering of the billion-level building surface and attribute data.

Description

Technical Field

[0001] This invention belongs to the field of housing data processing, specifically a method and system for visually assessing massive housing risks and hazards. Background Technology

[0002] In existing housing risk and disaster visualization systems, houses are presented as discrete building surface elements. These systems are used in applications such as housing censuses and cadastral surveys to establish the spatial location information of individual houses. However, at large spatial scales such as national, provincial, and municipal levels, users find it difficult to intuitively perceive the spatial distribution patterns. Furthermore, processing and rendering massive amounts of spatial data can easily lead to performance issues such as lag and memory exhaustion, affecting the page display. Therefore, housing attribute information is often displayed using statistical charts such as contour maps or bar charts to show the data distribution at different administrative levels, but this fails to reflect fine-grained spatial trends and cross-administrative regions.

[0003] Existing heat maps based on kernel density analysis are often used to display the spatial distribution characteristics of geographic elements. Their application is biased towards qualitative analysis and lacks interpretation of quantitative information. Therefore, it is difficult to obtain quantitative comparison results between heat maps of different times or spaces. For example, common heat maps only show the heat distribution, and users cannot obtain quantitative information through specific values, nor can they compare the changes in the amount of heat in a region at different time points. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for visually assessing massive housing risks and hazards, resolving technical issues such as lag and memory exhaustion during the rendering of hundreds of millions of building surfaces and attribute data.

[0005] The method of the present invention is achieved through the following technical solution: a method for visually assessing massive housing risks and hazards, comprising the following steps:

[0006] A comprehensive risk assessment of a single building is conducted by weighting and summing the multiple building attributes of the single building to calculate the risk hazard coefficient of the single building.

[0007] The massive number of houses are spatially aggregated to varying degrees at multiple administrative regions to reduce the overall time required for data analysis after aggregation and to obtain the distribution of risk hazards of houses at the aggregation points.

[0008] Kernel density analysis is used to convert discrete vector points into continuous raster data for display. The raster data is obtained by using the center coordinates of the raster cells in a specified row and column as the predicted points, setting corresponding search radii for different spatial scales at multiple administrative regions, and using kernel functions to perform weighted kernel density analysis on the predicted points to obtain raster data on the spatial distribution characteristics of renovation or demolition costs within a specified administrative region.

[0009] Based on the spatial distribution characteristics of heat grid data, the results are analyzed and visualized.

[0010] Preferably, the formula for calculating the risk hazard coefficient of a single-family house is as follows:

[0011]

[0012] ο represents the risk hazard coefficient of a single building, and N represents the total number of building attributes; a i To assign a value to the building attribute i, w i The hidden danger weight is the building attribute i.

[0013] More preferably, the cost of renovating or demolishing a single building is calculated based on the building risk hazard coefficient; the cost of renovating or demolishing a single building is calculated using a single-building renovation cost constructor, wherein the single-building renovation cost constructor is:

[0014]

[0015] Where o is the building risk coefficient, a is the area of ​​a single building, and f(o) is the cost required for building renovation or demolition.

[0016] After spatially aggregating a large number of houses at different levels across multiple administrative regions, the cost of renovating or demolishing the individual houses at the aggregation points is obtained.

[0017] Preferably, the cost of renovating or demolishing a single building is calculated based on the building risk hazard coefficient;

[0018] After spatially aggregating a large number of houses at multiple administrative levels to varying degrees, the costs for renovating or demolishing individual houses at the aggregation points are obtained; the spatial aggregation process includes:

[0019] Obtain the center point of the building map features and the corresponding renovation or demolition cost attribute. Based on the aggregation algorithm, reduce the number of original points within the administrative region level. Aggregate the original points according to their spatial distribution, sum the renovation costs of individual buildings of the original points included in the aggregation point, take the average value, and then assign it to the aggregation point as the renovation or demolition cost of the individual buildings of the aggregation point.

[0020] More preferably, the specific steps of spatial aggregation processing include:

[0021] K center points of building features are randomly selected and denoted as μ1, μ2, ..., μ k ;

[0022] Define the loss function:

[0023]

[0024] Where, x i c represents the original center point of the i-th house. i It is x i The cluster to which it belongs N represents the cluster center point corresponding to the cluster, and N is the number of house center points aggregated into this cluster;

[0025] Repeat the following process until the loss function J converges or reaches a set threshold to obtain the final cluster center points: x represents the original center point of each house. i Assigned to the nearest cluster center Recalculate the cluster center point for each cluster.

[0026] Calculate the summation of the average cost of modifying or demolishing the house center point for each cluster aggregation, and assign this value to the final generated cluster center point. Cost of renovation or demolition agg .

[0027] Preferably, the cost of renovating or demolishing a single building is calculated based on the building risk hazard coefficient;

[0028] After spatially aggregating a massive number of houses at multiple administrative region levels to varying degrees, the renovation or demolition costs of individual houses at the aggregated points are obtained. The spatial aggregation process yields the cluster centers. The process of converting discrete vector points into continuous raster data for visualization through kernel density analysis includes:

[0029] A grid is constructed using the coordinate range of the cluster center points after aggregation.

[0030] Traverse the pixels in the raster and search for all cluster centers whose distance from the pixel center is within the search radius h.

[0031] Calculate the distance d between the pixel center and the cluster center. p Based on the kernel function formula and the cost of renovating a single house at the cluster center point. agg The sum of the probability densities of the individual house renovation costs for all cluster center points relative to the predicted point is the predicted heat value for that cluster center point.

[0032] The system of the present invention is implemented through the following technical solution: a massive housing risk and hazard visualization assessment system, comprising the following modules:

[0033] The individual building risk assessment module is used to conduct a comprehensive building risk assessment of an individual building. It calculates the risk hazard coefficient of an individual building by weighted summing of multiple building attributes.

[0034] The unit cost conversion module calculates the cost of renovating or demolishing a single building based on the building risk hazard coefficient.

[0035] The aggregation module performs spatial aggregation processing on a massive number of houses at multiple administrative regions to varying degrees, thereby reducing the overall time required for data analysis after aggregation and obtaining the cost of individual house renovation or demolition at the aggregation point.

[0036] The kernel density analysis module converts discrete vector points into continuous raster data for display. The raster data is obtained by using the center coordinates of the specified row and column raster pixels as the predicted points, setting corresponding search radii for different spatial scales at multiple administrative regions, and using kernel functions to perform weighted kernel density analysis on the predicted points to obtain the spatial distribution characteristics of the heat of renovation or demolition costs within the specified administrative region.

[0037] The visualization output module performs result analysis and visualization rendering based on the grid data of heat spatial distribution characteristics.

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] 1. This invention transforms discrete building surface elements into continuous spatial trend surfaces, calculates heat distribution based on the density weight of the risk hazard index, and reflects the distribution patterns and changing trends of massive housing risk hazards at the national, provincial, and municipal scales. This enhances the display effect of the geographic spatial distribution characteristics of the housing risk hazard index, can efficiently process data and obtain display results, and solves technical problems such as lag and memory exhaustion when rendering hundreds of millions of building surfaces and attribute data.

[0040] 2. It achieves accurate geographic object mapping. Preprocessing of building feature points using the K-means algorithm, and comparing the kernel density analysis results of the aggregated points with the original points, shows that aggregation has little impact on the spatial distribution of kernel density and can still accurately reflect the spatial distribution of potentially hazardous buildings.

[0041] 3. It achieves efficient and reasonable heat calculation. A fourth-order function is selected as the kernel density function to balance the spatial continuity of kernel density analysis results with computational efficiency. Preprocessing by aggregating building element points effectively reduces the amount of data involved in kernel density calculations, improving analysis efficiency. Raster cells are evenly divided by row for multi-process processing, further enhancing data computation efficiency.

[0042] 4. It features intuitive and easy-to-understand visual results. The analysis and interpretation of quantitative information from kernel density analysis enhances users' understanding of the distribution of potentially hazardous buildings; the percentile calculation of grading color values ​​optimizes the visualization of heat maps. Attached Figure Description

[0043] Figure 1 This is a flowchart of the method for visually assessing the risks and hidden dangers of massive housing in this embodiment of the invention;

[0044] Figure 2 This is a schematic diagram of hierarchical rendering (grayscale display) within a provincial-level region in an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of hierarchical rendering (grayscale display) within a city-level area in an embodiment of the present invention. Detailed Implementation

[0046] The present invention will be further described below with reference to the accompanying drawings and embodiments, but the implementation of the present invention is not limited thereto.

[0047] Example 1

[0048] This embodiment provides a method for visually assessing massive housing risks and hazards, such as... Figure 1 As shown, it includes the following steps:

[0049] S1, Generate risk hazard coefficient for individual buildings

[0050] This step involves a comprehensive building risk assessment of a single building. Multiple building attributes, such as construction date, structural type, safety assessment level, whether it was professionally designed, and construction method, are weighted and summed to calculate the risk hazard coefficient (o) for the single building. The calculation formula used in this embodiment is as follows:

[0051]

[0052] ο represents the risk hazard coefficient of a single building, and N represents the total number of building attributes; a i To assign a value to the building attribute i, w i The hazard weights for building attribute i are shown in the table below:

[0053]

[0054]

[0055] S2. Calculate the cost of renovating or demolishing a single building based on the building risk hazard coefficient.

[0056] The risk coefficient 'o' calculated in step S1, combined with the condition that "houses with a risk coefficient of 4 need to be demolished and rebuilt, and individual houses with a risk coefficient of 2 or 3 need to be reinforced and renovated," is used to calculate the corresponding house renovation or demolition costs based on the individual house renovation cost constructor. The house individual renovation cost constructor is as follows:

[0057]

[0058] Where o is the building risk hazard coefficient, a is the area of ​​a single building, and f(o) is the cost required to renovate or demolish a building of this level.

[0059] This step further assesses the building risk factor based on the cost of renovating or demolishing a single building, so as to more specifically demonstrate the distribution of building risks in subsequent steps. However, those skilled in the art know that the distribution of building risks can also be demonstrated from other dimensions, such as the geographical range affected by high-risk buildings and the number of people living in high-risk buildings.

[0060] S3, Aggregate house points to reduce the number of original points.

[0061] After obtaining the renovation or demolition costs of individual houses, the houses can be aggregated before proceeding with further analysis. Due to the massive number of houses, spatial aggregation is performed to varying degrees at multiple administrative levels, such as national, provincial, and municipal. The results obtained from analyzing the data before and after aggregation are not significantly different, but the overall time required for analysis based on the aggregated data will be greatly reduced.

[0062] In this embodiment, the spatial aggregation process is as follows: The center point of the building map features and the corresponding renovation or demolition cost attribute are obtained. Based on the K-means point aggregation algorithm, the number of original points within the administrative region level is reduced (national-level clusters are set to 10,000, provincial-level clusters to 1,000, and municipal-level clusters to 100). The original points are aggregated according to their spatial distribution. The renovation costs of individual buildings encompassed by the aggregated points are summed and averaged, then assigned to the aggregated point as the renovation or demolition cost of the individual building at the aggregated point. The specific steps are as follows:

[0063] S31. Randomly select K center points of building map features, denoted as μ1, μ2, ..., μ k ;

[0064] S32. Define the loss function:

[0065]

[0066] Where, x i c represents the original center point of the i-th house. i It is x i The cluster to which it belongs N represents the cluster center point corresponding to the cluster, and N is the number of house center points aggregated into this cluster;

[0067] S33. Repeat the following two steps until the loss function J converges or reaches the set threshold to obtain the final cluster centers:

[0068] S331, Set the original center point x of each house i Assigned to the nearest cluster center

[0069] S332. Recalculate the cluster center point for each cluster.

[0070] S34. Calculate the summation of the average cost of modifying or demolishing the house center point for each cluster aggregation, and assign it as the final value for generating the cluster center point. Cost of renovation or demolition agg .

[0071] S4, kernel density analysis

[0072] After obtaining the cluster center points and the corresponding renovation or demolition costs for the corresponding building center points, the discrete vector points can be converted into continuous raster data for display using kernel density analysis. The raster data is obtained as follows: using the coordinates of the center points of specified row and column raster pixels as the predicted points, corresponding search radii (also known as search bandwidths, such as 100,000 meters for the national level, 50,000 meters for the provincial level, and 10,000 meters for the municipal level) are set for different spatial scales at multiple administrative regions. A fourth-order kernel function is used to perform weighted kernel density analysis on the predicted points to obtain raster data showing the spatial distribution characteristics of renovation or demolition costs within the specified administrative region.

[0073] In this embodiment, the nuclear density analysis process includes the following steps:

[0074] S41. Construct a grid using the coordinate range of the cluster center points after aggregation;

[0075] S42. Traverse the pixels in the raster and search for all cluster centers whose distance from the pixel center is within the search radius h.

[0076] S43. Calculate the distance d between the pixel center and the cluster center. pBased on the fourth-power kernel function formula and the cost of renovating a single house at the cluster center point... agg The sum of the probability densities of individual building renovation costs for all cluster center points relative to the predicted point is obtained, which is the predicted heat value f(px,py) for that cluster center point. The calculation formula is as follows:

[0077]

[0078]

[0079] Where M is the total number of prediction points, d p <h; px is the x-coordinate of the pixel center point, and py is the y-coordinate of the pixel center point; The x-coordinate of the cluster center point that was found during the search. The y-coordinate of the cluster center point that was found during the search.

[0080] In actual calculations, multi-processing technology is used to divide raster pixels into rows and enable multiple processes to process them in parallel, thereby improving the efficiency of kernel density analysis.

[0081] The calculation yields quantitative information on the pixel value: National level: predicted housing renovation costs within a 100,000-meter radius of the pixel; Provincial level: predicted housing renovation costs within a 50,000-meter radius of the pixel; Municipal level: predicted housing renovation costs within a 10,000-meter radius of the pixel.

[0082] S5. Based on the spatial distribution characteristics of heat grid data, perform result analysis and visualization rendering.

[0083] The generated thermal image pixel values ​​(i.e., raster data of thermal spatial distribution characteristics) are statistically analyzed. Then, a grading threshold for the pixel values ​​is calculated based on a percentile algorithm and used as the grading threshold for rendering. The thermal map is graded according to the 20%, 40%, 60%, and 80% percentile density values. Grading color bands are set to map different color values ​​to the corresponding grading ranges, improving the visualization effect. Figure 2 , Figure 3 As shown.

[0084] Example 2

[0085] Based on the same inventive concept as Embodiment 1, this embodiment provides a massive housing risk and hazard visualization assessment system, including the following modules:

[0086] The individual building risk assessment module is used to conduct a comprehensive building risk assessment of an individual building. It calculates the risk hazard coefficient of an individual building by weighted summing of multiple building attributes.

[0087] The unit cost conversion module calculates the cost of renovating or demolishing a single building based on the building risk hazard coefficient.

[0088] The aggregation module performs spatial aggregation processing on a massive number of houses at multiple administrative regions to varying degrees, thereby reducing the overall time required for data analysis after aggregation and obtaining the cost of individual house renovation or demolition at the aggregation point.

[0089] The kernel density analysis module converts discrete vector points into continuous raster data for display. The raster data is obtained by using the center coordinates of the specified row and column raster pixels as the predicted points, setting corresponding search radii for different spatial scales at multiple administrative regions, and using kernel functions to perform weighted kernel density analysis on the predicted points to obtain the spatial distribution characteristics of the heat of renovation or demolition costs within the specified administrative region.

[0090] The visualization output module performs result analysis and visualization rendering based on the grid data of heat spatial distribution characteristics.

[0091] The modules described above in this embodiment are used to execute the steps of embodiment 1, and the detailed execution process can be found in embodiment 1, which will not be repeated here.

[0092] The above description is only a preferred embodiment of the present invention, but the scope of protection of the invention patent is not limited thereto. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the scope of protection of the present invention.

Claims

1. A method for visually assessing massive housing risks and hazards, characterized in that, Includes the following steps: A comprehensive risk assessment of a single building is conducted by weighting and summing the multiple building attributes of the single building to calculate the risk hazard coefficient of the single building. The massive number of houses are spatially aggregated to varying degrees at multiple administrative regions to reduce the overall time required for data analysis after aggregation and to obtain the distribution of risk hazards of houses at the aggregation points. Kernel density analysis is used to convert discrete vector points into continuous raster data for display. The raster data is obtained by using the center coordinates of the raster cells in a specified row and column as the predicted points, setting corresponding search radii for different spatial scales at multiple administrative regions, and using kernel functions to perform weighted kernel density analysis on the predicted points to obtain raster data on the spatial distribution characteristics of renovation or demolition costs within a specified administrative region. Based on the spatial distribution characteristics of heat grid data, the results are analyzed and visualized. Calculate the cost of renovating or demolishing a single building based on the building risk hazard coefficient; After spatially aggregating a large number of houses at different levels of multiple administrative regions, the cost of renovating or demolishing the individual houses at the aggregation points is obtained. The spatial aggregation process includes: Obtain the center point of the building map features and the corresponding renovation or demolition cost attributes. Based on the aggregation algorithm, reduce the number of original points within the administrative region level. Aggregate the original points according to their spatial distribution. Sum the renovation costs of individual buildings of the original points included in the aggregation point and take the average value. Then assign the average value to the aggregation point as the renovation or demolition cost of individual buildings of the aggregation point. The specific steps of spatial aggregation processing include: Randomly select the center points of K house map features, denoted as ; Define the loss function: ； in, Represents the original center point of the i-th house. yes The cluster to which it belongs Represents the cluster center point corresponding to the cluster. The number of house center points aggregated into this cluster; Repeat the following process until the loss function is reached. Convergence, or reaching a set threshold, yields the final cluster center points: This involves converting the original center points of each house... Assigned to the nearest cluster center Recalculate the cluster center point for each cluster. ; Calculate the summation of the average cost of modifying or demolishing the house center point for each cluster aggregation, and assign this value to the final generated cluster center point. Cost of renovation or demolition .

2. The visualization evaluation method according to claim 1, characterized in that, The formula for calculating the risk factor of a detached house is: ； The risk factor for a single building. This represents the total number of building attributes. For the building attributes The assignment, For building attributes The weight of potential risks.

3. The visualization evaluation method according to claim 1 or 2, characterized in that, The cost of renovating or demolishing a single building is calculated based on the building risk hazard coefficient. The cost of renovating or demolishing a single building is calculated using a constructor for the cost of renovating a single building: ； Where 'o' represents the building risk hazard coefficient. This refers to the area of ​​a single building. Costs required for house renovation or demolition; After spatially aggregating a large number of houses at different levels across multiple administrative regions, the cost of renovating or demolishing the individual houses at the aggregation points is obtained.

4. The visualization evaluation method according to claim 1, characterized in that, The kernel function is a fourth-order kernel function.

5. The visualization evaluation method according to claim 1, characterized in that, Calculate the cost of renovating or demolishing a single building based on the building risk hazard coefficient; After spatially aggregating a large number of houses at different levels of multiple administrative regions, the cost of renovating or demolishing the individual houses at the aggregation points is obtained. Spatial aggregation processing is used to obtain the cluster center point after aggregation; The process of converting discrete vector points into continuous raster data for display using kernel density analysis includes: A grid is constructed using the coordinate range of the cluster center points after aggregation. Traverse the cells in the raster, searching for cells whose distance from the cell center is within the search radius. All cluster centers within; Calculate the distance between the pixel center and the cluster center. Based on the kernel function formula and the cost of renovating individual houses at the cluster center point The sum of the probability densities of the individual house renovation costs for all cluster center points relative to the predicted point is the predicted heat value for that cluster center point.

6. The visualization evaluation method according to claim 5, characterized in that, Heat forecast value The calculation formula is: ； ； in, The total number of predicted points, ;px is the x-coordinate of the pixel center point, and py is the y-coordinate of the pixel center point; The x-coordinate of the cluster center point that was found during the search. The y-coordinate of the cluster center point that was found during the search.

7. A massive housing risk and hazard visualization assessment system, characterized in that, Includes the following modules: The individual building risk assessment module is used to conduct a comprehensive building risk assessment of an individual building. It calculates the risk hazard coefficient of an individual building by weighted summing of multiple building attributes. The unit cost conversion module calculates the cost of renovating or demolishing a single building based on the building risk hazard coefficient. The aggregation module performs spatial aggregation processing on a massive number of houses at multiple administrative regions to varying degrees, thereby reducing the overall time required for data analysis after aggregation and obtaining the cost of individual house renovation or demolition at the aggregation point. The kernel density analysis module converts discrete vector points into continuous raster data for display. The raster data is obtained by using the center coordinates of the specified row and column raster pixels as the predicted points, setting corresponding search radii for different spatial scales at multiple administrative regions, and using kernel functions to perform weighted kernel density analysis on the predicted points to obtain the spatial distribution characteristics of the heat of renovation or demolition costs within the specified administrative region. The visualization output module performs result analysis and visualization rendering based on the grid data of heat spatial distribution characteristics; The spatial aggregation process performed by the aggregation module includes: The process involves obtaining the center point of the building map features and their corresponding renovation or demolition cost attributes. Based on an aggregation algorithm, the number of original points within the administrative region is reduced. These original points are then aggregated according to their spatial distribution. The renovation costs of individual buildings within the aggregated points are summed, averaged, and then assigned to the aggregated point as its individual building renovation or demolition cost. Specific steps include: Randomly select the center points of K house map features, denoted as ; Define the loss function: ； in, Represents the original center point of the i-th house. yes The cluster to which it belongs Represents the cluster center point corresponding to the cluster. The number of house center points aggregated into this cluster; Repeat the following process until the loss function is reached. Convergence, or reaching a set threshold, yields the final cluster center points: This involves converting the original center points of each house... Assigned to the nearest cluster center Recalculate the cluster center point for each cluster. ; Calculate the summation of the average cost of modifying or demolishing the house center point for each cluster aggregation, and assign this value to the final generated cluster center point. Cost of renovation or demolition .

Images