Building safety risk classification method and device and computer program product

Abstract

The invention relates to the technical field of building risk monitoring, in particular to a building safety risk classification method and device and a computer program product. According to the method, based on a permanent scatterer synthetic aperture radar interferometry technology, an SAR image of a monitoring area is calculated to generate InSAR deformation monitoring point cloud data; determining candidate PS points by unifying projection coordinate systems of building vector frame data and InSAR point cloud data and combining a vector frame buffer area and a measuring point height screening method; for the building of which the number of the candidate PS points reaches a preset threshold value, stripping seasonal fluctuation in the deformation data of the building, and extracting a nonlinear deformation trend; based on the nonlinear deformation trend, building a local coordinate system of the building and establishing a differential settlement model, calculating differential settlement indexes of the building in the X and Y directions, and generating a time sequence; by extracting indexes such as the maximum value of the time sequence, the average change rate and the over-limit time proportion, safety risk classification is carried out on the building, and the accuracy of building safety risk classification is improved.

Description

Technical Field

[0001] The present application relates to the technical field of building risk monitoring, and in particular to a method, device and computer program product for classifying building safety risks. Background Art

[0002] With the acceleration of urbanization, a large number of existing buildings, especially old buildings, are affected by geological conditions, environmental disturbances and climate change during their service, and are prone to settlement and structural deformation, and even cause safety accidents. Therefore, building safety monitoring and risk assessment have become important tasks in urban management. Existing monitoring methods such as total station observation, ground sensors and laser plumb meters can provide certain monitoring effects, but they have obvious limitations in large-scale building screening and automation. These methods are subject to large environmental restrictions and high labor input, and cannot meet the needs of efficient monitoring of a large number of buildings in modern cities. Synthetic aperture radar interferometry (InSAR) technology, especially permanent scatterer InSAR (PS-InSAR) technology, has become an important tool for building deformation monitoring due to its wide coverage, high precision and time series monitoring advantages. However, when applied to building monitoring, existing InSAR technology faces problems such as insufficient PS point screening accuracy, seasonal fluctuation interference and building settlement feature modeling, resulting in poor accuracy in building safety risk classification. Therefore, how to improve the accuracy of building safety risk classification has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The main purpose of this application is to provide a building safety risk classification method, device and computer program product, aiming to solve the technical problem of how to improve the accuracy of building safety risk classification.

[0004] To achieve the above objectives, the present application provides a method for classifying building safety risks, the method comprising the following steps:

[0005] Based on the permanent scatterer synthetic aperture radar interferometry technology, the SAR image of the monitoring area is solved to generate InSAR deformation monitoring point cloud data;

[0006] Unify the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determine the candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method;

[0007] If the number of the candidate PS points is not less than a preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuation removed;

[0008] Based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established according to the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model, and a time series of the differential settlement index is generated;

[0009] Characteristic indicators are extracted from the time series of the differential settlement index, and based on the characteristic indicators, the buildings in the monitoring area are classified for safety risks; wherein the characteristic indicators include the maximum value of the time series, the average rate of change, and the proportion of over-limit time.

[0010] In one embodiment, before the step of solving the SAR image of the monitoring area based on the permanent scatterer synthetic aperture radar interferometry technology to generate InSAR deformation monitoring point cloud data, the step further includes:

[0011] Based on the location data of the monitoring area, an initial SAR image is acquired from a preset satellite remote sensing data platform;

[0012] Based on a preset time baseline and a preset space baseline, the initial SAR image is screened to obtain a screened SAR image;

[0013] The filtered SAR image is preprocessed to obtain the SAR image; wherein the preprocessing includes one or more of radiation correction, geometric correction and registration.

[0014] In one embodiment, the step of solving the SAR image of the monitoring area based on the permanent scatterer synthetic aperture radar interferometry technology to generate InSAR deformation monitoring point cloud data includes:

[0015] Based on the permanent scatterer synthetic aperture radar interferometry technology, the SAR images are paired to generate an interferogram sequence;

[0016] Determining an initial PS point according to the phase consistency and scattering intensity of the interference patterns in the interference pattern sequence;

[0017] De-noising the phase time series of the initial PS point to obtain interference phases at multiple time points;

[0018] Based on the phase unwrapping algorithm, the interference phases of the multiple time points are continuously solved to determine the accumulated deformation information and height information of the initial PS point;

[0019] The geographic coordinate information of the initial PS point, the accumulated deformation information and the height information are aggregated to generate the InSAR deformation monitoring point cloud data.

[0020] In one embodiment, the step of unifying the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determining the candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method, comprises:

[0021] Converting the coordinate system of the building vector frame data into the projection coordinate system of the InSAR deformation monitoring point cloud data;

[0022] A buffer zone of a preset distance is set at the boundary of each building vector frame, and based on the building vector frame range and the buffer zone, the initial PS points in the InSAR deformation monitoring point cloud data are spatially screened to obtain spatially screened PS points;

[0023] The estimated elevation of the spatial screening PS point is compared with the building height range, and based on the comparison result, the spatial screening PS point is height screened to determine the candidate PS point.

[0024] In one embodiment, before the step of stripping seasonal fluctuation components from the deformation data of the candidate PS points to obtain a nonlinear deformation trend without seasonal fluctuations if the number of the candidate PS points is not less than a preset number threshold, the method further includes:

[0025] Comparing the number of the candidate PS points with the preset number threshold;

[0026] If the number of the candidate PS points is less than the preset number threshold, the corresponding building is marked as an uncomputable building;

[0027] The maximum cumulative sedimentation value, long-term sedimentation rate and short-term sedimentation rate of the candidate PS point are obtained as auxiliary data for manual analysis.

[0028] In one embodiment, the step of extracting characteristic indicators from the time series of the differential settlement index and classifying the safety risks of the buildings in the monitoring area based on the characteristic indicators includes:

[0029] Performing data cleaning and data completion on the time series of the differential sedimentation index to obtain a target sequence;

[0030] Obtaining the maximum absolute value of the differential sedimentation index in the target sequence as the maximum value of the time series;

[0031] Determining an average rate of change based on the total time span of the target sequence and the total change in differential sedimentation index;

[0032] According to the preset risk threshold, the total time of exceeding the limit is determined, and the ratio of the total time of exceeding the limit to the total time of the target sequence is used as the time ratio of exceeding the limit;

[0033] The maximum value of the time series is compared with a preset maximum value threshold, the average change rate is compared with a preset rate threshold, and the over-limit time ratio is compared with a preset over-limit ratio threshold, and the risk classification is determined based on the comparison results.

[0034] In one embodiment, the step of comparing the time series maximum value with a preset maximum value threshold, the average change rate with a preset rate threshold, and the over-limit time ratio with a preset over-limit ratio threshold, and determining the risk classification according to the comparison results includes:

[0035] Compare the maximum value of the time series with the preset maximum value threshold, and if the maximum value of the time series is greater than the preset maximum value threshold, mark the maximum value of the time series as a risk indicator;

[0036] Comparing the average change rate with the preset rate threshold, and if the average change rate is greater than the preset rate threshold, marking the average change rate as the risk indicator;

[0037] Comparing the over-limit time ratio with the preset over-limit ratio threshold, and if the over-limit time ratio is greater than the preset over-limit ratio threshold, marking the over-limit time ratio as the risk indicator;

[0038] Based on the number of risk indicators, a risk classification is determined.

[0039] In one embodiment, if the number of the candidate PS points is not less than a preset number threshold, the step of stripping the seasonal fluctuation component in the deformation data of the candidate PS points to obtain a nonlinear deformation trend with seasonal fluctuation removed includes:

[0040] If the number of the candidate PS points is not less than a preset number threshold, extracting deformation data of the candidate PS points;

[0041] Based on the deformation data, a time series is generated, and the time series is aligned;

[0042] Based on a preset time series analysis algorithm, detecting the seasonal fluctuation component in the time series;

[0043] Based on the fluctuation period of the seasonal fluctuation component, a fitting algorithm is used to remove the seasonal fluctuation component from the time series to obtain the nonlinear deformation trend of removing the seasonal fluctuation.

[0044] In addition, to achieve the above purpose, the present application also proposes a building safety risk classification device, the building safety risk classification device comprising:

[0045] Image solving module, used to solve the SAR image of the monitoring area based on the permanent scatterer synthetic aperture radar interferometry technology to generate InSAR deformation monitoring point cloud data;

[0046] A candidate point confirmation module is used to unify the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determine the candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method;

[0047] A fluctuation stripping module, for stripping seasonal fluctuation components from the deformation data of the candidate PS points if the number of the candidate PS points is not less than a preset number threshold, to obtain a nonlinear deformation trend with seasonal fluctuations removed;

[0048] A differential settlement module is used to construct a local building coordinate system based on the nonlinear deformation trend with seasonal fluctuations removed, and to establish a building differential settlement model according to the local building coordinate system, to determine the differential settlement index of the building in the X and Y directions through the building differential settlement model, and to generate a time series of the differential settlement index;

[0049] The target module is used to extract characteristic indicators from the time series of the differential settlement index, and classify the safety risks of the buildings in the monitoring area based on the characteristic indicators; wherein the characteristic indicators include the maximum value of the time series, the average change rate and the proportion of over-limit time.

[0050] In addition, to achieve the above-mentioned purpose, the present application also proposes a building safety risk classification device, which includes: a memory, a processor, and a building safety risk classification program stored on the memory and executable on the processor, and the building safety risk classification program is configured to implement the steps of the building safety risk classification method described above.

[0051] In addition, to achieve the above-mentioned purpose, the present application also proposes a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the building safety risk classification method as described above.

[0052] The present application is based on the permanent scatterer synthetic aperture radar interferometry technology to solve the SAR image of the monitoring area and generate InSAR deformation monitoring point cloud data; unify the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determine the candidate PS points by setting a buffer zone at the edge of the vector frame and the measuring point height screening method; if the number of candidate PS points is not less than a preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuations removed; based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established based on the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model to generate a time series of differential settlement indexes; characteristic indicators are extracted from the time series of differential settlement indexes, and based on the characteristic indicators, the safety risk classification of the buildings in the monitoring area is performed; wherein the characteristic indicators include the maximum value of the time series, the average change rate, and the proportion of over-limit time. This application improves the accuracy of building safety risk classification by utilizing InSAR deformation monitoring point cloud data, accurate building vector frame and PS point screening, stripping off seasonal fluctuations, establishing a building differential settlement model, and extracting key characteristic indicators; InSAR technology provides high-precision long-term settlement data, building vector frame and PS point screening ensure the pertinence of the data, stripping off seasonal fluctuations removes short-term environmental impacts, and the differential settlement model accurately reflects the settlement characteristics of the building in the X and Y directions. Finally, classification is performed based on multi-dimensional characteristic indicators such as the maximum value of the time series, the average change rate, and the excess time ratio, which comprehensively reflects the building settlement risk characteristics, thereby improving the accuracy of building safety risk classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of the first embodiment of the building safety risk classification method of the present application;

[0054] Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the building safety risk classification method of this application;

[0055] Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the building safety risk classification method of this application;

[0056] Figure 4 This is a schematic diagram of a local building coordinate system in an embodiment of the building safety risk classification method of the present application;

[0057] Figure 5 This is a schematic diagram of a building overall differential settlement model in an embodiment of a building safety risk classification method of the present application;

[0058] Figure 6This is a schematic diagram of the module structure of the building safety risk classification device according to an embodiment of the present application;

[0059] Figure 7 Schematic diagram of the equipment structure of the hardware operating environment involved in the building safety risk classification method in the embodiment of the present application.

[0060] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0061] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0062] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0063] It should be noted that with the acceleration of urbanization, a large number of existing buildings, especially old buildings, are affected by geological conditions, environmental disturbances and climate change during their service, and are prone to settlement and structural deformation, and even cause safety accidents. Therefore, building safety monitoring and risk assessment have become important tasks in urban management. Existing monitoring methods such as total station observation, ground sensors and laser plumb meters can provide certain monitoring effects, but they have obvious limitations in large-scale building screening and automation. These methods are subject to greater environmental restrictions and high labor input, and cannot meet the needs of efficient monitoring of a large number of buildings in modern cities. Synthetic aperture radar interferometry (InSAR) technology, especially permanent scatterer InSAR (PS-InSAR) technology, has become an important tool for building deformation monitoring due to its wide coverage, high precision and time series monitoring advantages. However, when applied to building monitoring, existing InSAR technology faces problems such as insufficient PS point screening accuracy, seasonal fluctuation interference and building settlement feature modeling, resulting in poor accuracy in building safety risk classification. Therefore, how to improve the accuracy of building safety risk classification has become a technical problem that needs to be solved urgently.

[0064] The main solution of the present application is: based on the permanent scatterer synthetic aperture radar interferometry technology, the SAR image of the monitoring area is solved to generate InSAR deformation monitoring point cloud data; the building vector frame data of the monitoring area is unified with the projection coordinate system of the InSAR deformation monitoring point cloud data, and the candidate PS points are determined by setting a buffer zone at the edge of the vector frame and the measuring point height screening method; if the number of candidate PS points is not less than the preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuations removed; based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established according to the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model, and a time series of differential settlement index is generated; characteristic indicators are extracted from the time series of differential settlement index, and based on the characteristic indicators, the safety risk classification of the buildings in the monitoring area is performed; wherein the characteristic indicators include the maximum value of the time series, the average rate of change, and the proportion of over-limit time.

[0065] This application improves the accuracy of building safety risk classification by utilizing InSAR deformation monitoring point cloud data, accurate building vector frame and PS point screening, stripping off seasonal fluctuations, establishing a building differential settlement model, and extracting key characteristic indicators; InSAR technology provides high-precision long-term settlement data, building vector frame and PS point screening ensure the pertinence of the data, stripping off seasonal fluctuations removes short-term environmental impacts, and the differential settlement model accurately reflects the settlement characteristics of the building in the X and Y directions. Finally, classification is performed based on multi-dimensional characteristic indicators such as the maximum value of the time series, the average change rate, and the excess time ratio, which comprehensively reflects the building settlement risk characteristics, thereby improving the accuracy of building safety risk classification.

[0066] It should be noted that the execution subject of the method of this embodiment can be a computing service device with data processing, network communication and program running functions, or it can be the above-mentioned building safety risk classification device with the same or similar functions. This embodiment and the following embodiments will be described by taking the building safety risk classification device as an example.

[0067] Based on this, the first embodiment of the building safety risk classification method of this application is proposed, please refer to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the building safety risk classification method of the present application.

[0068] In this embodiment, the building safety risk classification method includes the following steps:

[0069] S1: Based on the permanent scatterer synthetic aperture radar interferometry technology, the SAR image of the monitoring area is solved to generate InSAR deformation monitoring point cloud data;

[0070] It should be noted that the Permanent Scatter Synthetic Aperture Radar Interferometry (PS-InSAR) is a surface monitoring technology based on radar images. It uses long-term stable high-coherence reflection points (permanent scatterers, PS points) on the ground to monitor tiny surface deformations. PS-InSAR identifies these PS points in the time series and obtains their cumulative settlement information and height information by analyzing their phase changes. Synthetic Aperture Radar (SAR) is an active remote sensing technology that generates surface images by transmitting and receiving microwave signals. It has the advantages of all-weather and not being restricted by lighting conditions, and is suitable for large-scale surface deformation monitoring. Interferometry (InSAR) is a measurement technology that obtains its relative height or displacement information by comparing the phase difference of the same surface point in two or more SAR images. InSAR is particularly suitable for wide-area surface deformation monitoring. Point cloud data is a form of spatial data, consisting of the coordinates of multiple points (such as X, Y, and Z coordinates) that reflect the spatial position. The deformation monitoring point cloud data generated by InSAR contains the position, height, and deformation information of the permanent scatterers on the surface.

[0071] Specifically, multi-phase SAR images covering the monitoring area are obtained from the satellite remote sensing data platform to ensure that the images are collected at different time points to reflect the surface deformation trend of the monitoring area over a long period of time. The acquisition of SAR images must meet certain time and space baselines to ensure sufficient coherence. In addition, in order to reduce the radiation errors of the sensor and the environment, the SAR images are subjected to radiation correction and geometric correction to ensure the radiation consistency and geometric accuracy of each image.

[0072] Furthermore, the preprocessed SAR images were analyzed using PS-InSAR technology. Through phase consistency analysis of the time series, permanent scatterers (PS points) that maintained high coherence in multiple imaging were identified. The phase change information of these PS points represents the surface displacement or settlement. Subsequently, the cumulative settlement and height of each PS point were calculated to generate InSAR deformation monitoring point cloud data containing these PS points. This point cloud data provides deformation information of the surface in the monitoring area, which is an important basis for subsequent building deformation analysis and risk classification.

[0073] The InSAR deformation monitoring point cloud data generated by this step can achieve high-precision measurement of surface deformation in the monitoring area. PS-InSAR technology relies on highly coherent permanent scatterers to overcome the problems of traditional surface observation methods being limited by environmental conditions and insufficient monitoring accuracy. Its high spatial resolution can capture tiny settlement changes in buildings, and long-term SAR image analysis ensures the stability and accuracy of monitoring. This method can not only capture surface deformation over a large range and around the clock, but also provide high-precision PS point deformation data, forming a stable and reliable settlement monitoring basis for buildings and surrounding areas, and effectively improving the accuracy of building safety risk classification.

[0074] S2: unifying the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determining candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method;

[0075] It should be noted that the building vector frame data is the boundary data that represents the spatial position of the building in vector form, and usually contains the geographic coordinate information of the building's outline and corner points. These data can help locate the spatial position of the building and are used to screen the PS points on the surface of the building in the InSAR deformation monitoring point cloud. The projection coordinate system is a method of mapping the three-dimensional spatial position of the earth's surface to a two-dimensional plane. Different data may use different projection coordinate systems. In order to ensure the spatial matching of the data, different data need to be converted to the same projection coordinate system. The buffer zone refers to an additional spatial area set at the edge of the vector box, which is usually extended at a certain distance to include PS points near the edge of the building. The setting of the buffer zone can cover stable points around the building and improve the representativeness of the data. The height screening of the measuring point is to use the screening criteria of the building height range to identify PS points that match the height of the building structure to ensure that these PS point data accurately reflect the settlement of the building.

[0076] Specifically, the building vector frame data and InSAR deformation monitoring point cloud data are converted to the same projection coordinate system. Building vector frame data usually uses a geographic coordinate system (such as WGS 84), while InSAR data may use a projection coordinate system (such as UTM). After the coordinate system is unified, the spatial position of the building can be directly matched with the position of the InSAR point cloud data, laying the foundation for the subsequent accurate screening of candidate PS points.

[0077] Furthermore, a buffer zone of a preset distance is set at the edge of each building vector box to include PS points near the building edge. Then, the PS points within the vector box and its buffer zone are screened out through a spatial screening method. These screened PS points are further screened for height, and the estimated height of the PS points is compared with the height range of the building to eliminate PS points that do not meet the building height. In this way, a set of candidate PS points related to the building structure is finally determined, providing accurate input data for the next step of settlement analysis.

[0078] By unifying the projection coordinate system and accurately screening candidate PS points, the PS point data used for analysis is ensured to accurately match the spatial position of the building. Setting a buffer zone can effectively cover the PS points at the edge of the building, and combining height screening to further eliminate irrelevant points makes the final candidate PS point set more representative and targeted. This precise PS point screening method improves data quality and analysis accuracy, and can more realistically reflect the settlement and deformation of the building, thereby providing a reliable data basis for the accurate classification of building safety risks.

[0079] S3: if the number of the candidate PS points is not less than a preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuation removed;

[0080] Step S3 includes:

[0081] S31: if the number of the candidate PS points is not less than a preset number threshold, extracting deformation data of the candidate PS points;

[0082] S32: generating a time series based on the deformation data, and aligning the time series;

[0083] S33: Detecting the seasonal fluctuation component in the time series based on a preset time series analysis algorithm;

[0084] S34: Based on the fluctuation period of the seasonal fluctuation component, a fitting algorithm is used to separate the seasonal fluctuation component from the time series to obtain the nonlinear deformation trend of removing the seasonal fluctuation.

[0085] It should be noted that the candidate PS points are permanent scatterer points closely related to the building structure, which are screened out from the InSAR deformation monitoring point cloud data through the building vector box, buffer range and height screening method. These PS points represent stable reflection points on the surface or structure of the building, and can provide accurate information on the settlement and deformation of the building. The seasonal fluctuation component is a periodic change caused by seasonal factors such as temperature and humidity. This fluctuation will cause short-term periodic fluctuations in the deformation data of the PS points, which may mask the true long-term settlement trend of the building. The nonlinear deformation trend refers to the deformation trend that remains after removing the seasonal fluctuation, which is more representative of the actual settlement or deformation of the building during the monitoring period. This nonlinear trend is a smoother change sequence that can more accurately reflect the long-term structural deformation of the building.

[0086] Specifically, it is determined whether the number of candidate PS points screened out meets the preset number threshold requirement. The threshold ensures that the number of PS points used for analysis is sufficient to support the subsequent deformation trend analysis. If the number of candidate PS points reaches the threshold, it means that the data volume is sufficient and further seasonal fluctuation stripping analysis can be performed. It should be noted that if the number of measurement points is insufficient, the differential settlement index cannot be calculated.

[0087] Furthermore, the seasonal fluctuation components are stripped from the time series deformation data of the candidate PS points. The seasonal fluctuation components in the deformation data are identified and separated through periodic fluctuation detection methods (such as Fourier analysis or wavelet decomposition) and then stripped. The stripped data mainly retains the long-term, non-periodic deformation components, that is, the nonlinear settlement trend of the building. This nonlinear deformation trend more realistically reflects the long-term structural changes of the building and provides more reliable data for subsequent safety risk classification.

[0088] By stripping off the seasonal fluctuation component, the short-term fluctuation interference caused by cyclical changes in climate or environment is removed, and a more stable nonlinear deformation trend is obtained. This process makes the long-term settlement trend of the building clearer and effectively prevents short-term fluctuations from misleading the settlement data. This nonlinear trend can more accurately reflect the actual settlement state of the building, providing more accurate input data for the safety risk classification of the building, thereby improving the accuracy of risk assessment.

[0089] S4: Based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established according to the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model, and a time series of the differential settlement index is generated;

[0090] It should be noted that the local coordinate system of the building is a local reference coordinate system constructed with the building as the center. By adjusting the direction of the coordinate axis to be consistent with the main structural direction of the building (such as the long and short axes of the building), it is convenient to accurately describe the settlement characteristics of the building in the X and Y directions. The building differential settlement model is a mathematical model that calculates the settlement differences of the building in different directions by analyzing the nonlinear deformation trend of the PS point in the building. This model can quantify the settlement distribution of the building in the horizontal plane and reflect the tilt or uneven settlement of the building. The differential settlement index refers to the quantitative index of the settlement difference of the building in the X and Y directions. It is calculated based on the local coordinate system of the building and the differential settlement model, and is usually expressed in the form of a time series to reflect the settlement characteristics of the building over time.

[0091] Specifically, based on the nonlinear deformation trend obtained after stripping off seasonal fluctuations, a local coordinate system of the building is constructed. The long and short axis directions of the building are determined by the geometric information of the building vector frame, and a PS point on the building is used as the reference point (such as a center point of the PS point concentration area). Subsequently, the coordinate axes in the WE and NS directions are adjusted to be consistent with the long and short axis directions of the building, thereby defining the local coordinate system of the building. This local coordinate system can more accurately describe the spatial deformation characteristics of the building.

[0092] Furthermore, the building local coordinate system is used to establish a building differential settlement model. By analyzing the nonlinear deformation trend of the PS point in the building, the relative settlement difference of the PS point in the X and Y directions is calculated, and the uneven settlement of the building in two directions is quantified. The specific calculation includes: determining the projection of the PS point in the local coordinate system according to its spatial position; calculating the differential settlement value in each direction using the geometric relationship of the overall settlement of the building; and generating the differential settlement index of the building in the X and Y directions by integrating the differential settlement information of the PS point. The differential settlement index is arranged in time to form a time series of the differential settlement index, which provides dynamic data for subsequent building safety risk assessment.

[0093] By constructing a local building coordinate system and a differential settlement model, the settlement characteristics of a building in the horizontal plane can be described more accurately. This method avoids the errors that may be introduced by calculations in a geographic coordinate system and improves the ability to quantify building tilt or uneven settlement. The time series of the differential settlement index provides dynamic information on building settlement over time, helps capture abnormal changes in the building settlement process, and provides high-precision input data for building safety risk classification, thereby enhancing the reliability and effectiveness of risk assessment.

[0094] S5: extracting characteristic indicators from the time series of the differential settlement index, and classifying the buildings in the monitoring area for safety risks based on the characteristic indicators; wherein the characteristic indicators include the maximum value of the time series, the average rate of change, and the proportion of over-limit time;

[0095] It should be noted that the time series of the differential settlement index is the settlement difference index of the building in the X and Y directions calculated based on the local coordinate system of the building and the differential settlement model, and is arranged in chronological order to form a time series for dynamic monitoring of the settlement characteristics of the building. The maximum value of the time series refers to the maximum settlement difference that appears in the differential settlement index time series, which indicates the value of the maximum uneven settlement index of the building during the monitoring period. The average rate of change is the average change of the differential settlement index over time, reflecting the long-term trend and speed of uneven settlement of the building. The excess time ratio refers to the proportion of time points in the time series when the differential settlement index exceeds the preset safety threshold to the total time points, which measures the length of time the building is in a risky state during the monitoring period.

[0096] Specifically, the time series of differential sedimentation index is preprocessed, including data cleaning (removing outliers and noise) and data completion (filling missing values ​​through interpolation methods). The preprocessed time series is more stable and complete, providing a reliable input for feature index extraction.

[0097] Furthermore, the following characteristic indicators are extracted from the time series: Maximum value of the time series: directly search for the maximum value in the time series, record its value and the corresponding time point, and the maximum value represents the most serious settlement difference of the building during the monitoring period; Average change rate: calculate the total time span of the time series and the total change of the settlement index, and use the formula average change rate = total change / total time span to quantify the uneven settlement speed of the building; Exceeding time ratio: set a safety threshold, compare the time series point by point, count the number of time points exceeding the threshold, and calculate its proportion of the total number of time points to measure the risk exposure time of the building. Finally, based on the above characteristic indicators, the safety risk of the building is classified according to the preset classification rules. For example: if all characteristic indicators are lower than the set threshold, it is classified as low risk; if one or two indicators are close to or exceed the threshold, it is classified as medium risk; if multiple indicators obviously exceed the threshold, it is classified as high risk.

[0098] By extracting characteristic indicators from the time series of differential settlement index and classifying risks based on these indicators, the settlement characteristics and risk status of buildings can be fully and accurately reflected. The maximum value of the time series quantifies the most serious settlement state of the building, the average rate of change captures the settlement trend, and the proportion of overlimit time measures the extent to which the building is in an unsafe state for a long time. The combination of these indicators makes risk classification more scientific and detailed, can effectively identify building safety hazards, provide managers with a clear decision-making basis, and improve the efficiency and accuracy of urban building safety monitoring.

[0099] This embodiment is based on the permanent scatterer synthetic aperture radar interferometry technology to solve the SAR image of the monitoring area and generate InSAR deformation monitoring point cloud data; the building vector frame data of the monitoring area is unified with the projection coordinate system of the InSAR deformation monitoring point cloud data, and the candidate PS points are determined by setting a buffer zone at the edge of the vector frame and the vector frame range and the measurement point height screening method; if the number of candidate PS points is not less than the preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuations removed; based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established according to the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model, and a time series of the differential settlement index is generated; characteristic indicators are extracted from the time series of the differential settlement index, and the safety risk classification of the buildings in the monitoring area is performed based on the characteristic indicators; wherein the characteristic indicators include the maximum value of the time series, the average change rate, and the over-limit time ratio. This embodiment improves the accuracy of building safety risk classification by utilizing InSAR deformation monitoring point cloud data, accurate building vector frame and PS point screening, stripping seasonal fluctuations, establishing a building differential settlement model, and extracting key characteristic indicators; InSAR technology provides high-precision long-term settlement data, building vector frame and PS point screening ensure the pertinence of the data, stripping seasonal fluctuations removes short-term environmental impacts, and the differential settlement model accurately reflects the settlement characteristics of the building in the X and Y directions. Finally, classification is performed based on multi-dimensional characteristic indicators such as the maximum value of the time series, the average change rate, and the excess time ratio, which comprehensively reflects the building settlement risk characteristics, thereby improving the accuracy of building safety risk classification.

[0100] Based on the above first embodiment, a second embodiment of the building safety risk classification method of the present application is proposed. Figure 2 , Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the building safety risk classification method of the present application.

[0101] like Figure 2 As shown, in this embodiment, before step S1, the following is further included:

[0102] S1a: Based on the location data of the monitoring area, an initial SAR image is obtained from a preset satellite remote sensing data platform;

[0103] S1b: Based on a preset time baseline and a preset space baseline, the initial SAR image is screened to obtain a screened SAR image;

[0104] S1c: Preprocessing the filtered SAR image to obtain the SAR image; wherein the preprocessing includes one or more of radiation correction, geometric correction and registration.

[0105] It should be noted that the initial SAR image is a synthetic aperture radar image covering the monitoring area obtained through a satellite remote sensing data platform, which serves as the basic data source for InSAR analysis. The time baseline refers to the time interval between two SAR images. The time baseline determines the time resolution of deformation in the monitoring area. Reasonable time baseline selection can balance the frequency and accuracy of deformation detection. The spatial baseline refers to the orbital spacing between satellites during two imagings, which affects the interferometric quality of the image and the accuracy of height measurement. Too large or too small a spatial baseline will reduce the coherence of the interferogram. Radiation correction refers to the brightness consistency processing of SAR images to eliminate the influence of sensor characteristics and imaging environment on radar signals and ensure the accuracy of the radiation intensity of the image. Geometric correction refers to correcting the geometric distortion of SAR images to make the spatial position of the image consistent with the geographic coordinates. Registration refers to the spatial alignment of multi-phase SAR images to ensure that their pixel positions are consistent under the same spatial reference for interferometric analysis.

[0106] Specifically, according to the location data of the monitoring area, multi-temporal SAR images covering the area are downloaded from the preset satellite remote sensing data platform (such as Sentinel-1, TerraSAR-X or Radarsat, etc.). The selected images should have high spatial resolution and temporal coverage to meet the needs of surface deformation monitoring. The acquisition of initial SAR images needs to comprehensively consider factors such as the terrain characteristics of the monitoring area, satellite orbit parameters, and imaging angles. Among the downloaded initial SAR images, SAR images with moderate time intervals are selected to avoid insufficient monitoring deformation due to too short time intervals, or decreased image coherence due to too long intervals. Images with moderate spatial orbital spacing are screened to ensure that the interferometric phase has good coherence and improve the accuracy of interferometric measurement. The screened SAR images are used as input for subsequent processing to ensure the quality and availability of monitoring data.

[0107] Furthermore, the selected SAR images are preprocessed by radiation correction, geometric correction and registration: Radiation correction: eliminate the brightness difference caused by sensor characteristics or external environment during radar imaging to ensure the physical meaning of image brightness is consistent; Geometric correction: correct the geometric distortion of SAR images and accurately match the pixel position of the image with the geographic coordinates for subsequent spatial alignment with the building vector frame data; Registration: align the multi-phase SAR images to the same spatial reference system to ensure that the data at each time point in the subsequent analysis can be accurately compared. Pixel misalignment and spatial errors are reduced through fine registration. Finally, the preprocessed SAR images have the characteristics of radiation consistency, geometric accuracy and time series alignment, providing high-quality data input for subsequent InSAR analysis.

[0108] By obtaining the initial SAR images from the satellite remote sensing platform and screening and preprocessing them in time and space baselines, the high quality and coherence of the input data are ensured. The reasonable screening of time and space baselines improves the time resolution and spatial resolution of deformation monitoring; radiation correction enhances the physical consistency of the image, geometric correction ensures the accurate matching of the image and geographic coordinates, and registration eliminates the spatial errors of multi-phase images. This series of processing provides stable and reliable basic data for the subsequent generation of InSAR point clouds and building safety risk assessment, significantly improving the monitoring accuracy and reliability of data analysis.

[0109] Based on the above first embodiment, in this embodiment, step S1 includes:

[0110] S11: Based on the permanent scatterer synthetic aperture radar interferometry technology, performing image pairing on the SAR images to generate an interferogram sequence;

[0111] S12: determining an initial PS point according to the phase consistency and scattering intensity of the interference pattern in the interference pattern sequence;

[0112] S13: performing denoising processing on the phase time series of the initial PS point to obtain interference phases at multiple time points;

[0113] S14: Based on a phase unwrapping algorithm, continuously solve the interference phases of the multiple time points to determine the accumulated deformation information and height information of the initial PS point;

[0114] S15: Summarize the geographic coordinate information of the initial PS point, the accumulated deformation information, and the height information to generate the InSAR deformation monitoring point cloud data.

[0115] It should be noted that image pairing refers to combining SAR images taken at different times in the same area to generate an interferogram for measuring surface deformation. Image pairing can capture the relative displacement information of surface points within a time interval. An interferogram is an image generated by pairing the phase difference of two SAR images. The interferogram reflects the surface deformation and terrain undulation information and is the basic data for InSAR analysis. Phase consistency refers to the fact that certain surface scattering points maintain stable consistency in phase during the interferometric processing of multi-temporal images. These points are highly coherent permanent scatterers (PS points). Permanent scatterers (PS) points refer to stable scatterers that maintain high coherence and do not change over time in multiple SAR images. They are usually located on non-deformable objects such as buildings and bridges, and are suitable for accurate monitoring of surface deformation. Denoising refers to filtering or smoothing the phase time series of PS points to reduce noise interference and improve the accuracy of phase data. Phase unwrapping is the process of extracting continuous displacement or height information from the phase of the interferogram. By eliminating blur, the accumulated settlement information and height information of the PS points are obtained. The InSAR deformation monitoring point cloud data contains a collection of information such as the geographical location, cumulative deformation, and height of the PS point, which is used for subsequent building settlement analysis.

[0116] Specifically, the multi-temporal SAR images of the monitoring area are paired, and the SAR images of the same surface area at different times are combined to generate multiple interferograms. These interferograms record the phase difference information between the two imaging time points of the monitoring area and contain quantitative data of the surface micro-deformation. Through multi-temporal pairing, a continuous interferogram sequence is formed to provide time series data for subsequent PS point identification.

[0117] Furthermore, according to the generated interferogram sequence, the phase consistency and scattering intensity of the surface points in each interferogram are analyzed to determine the high coherence and stable initial PS points. These PS points maintain high consistency in the time series and are suitable for long-term deformation monitoring. Next, the phase time series of the initial PS points is denoised to reduce the interference of environmental and sensor noise and obtain purer phase data. Then, the denoised phase data is continuously solved using the phase unwrapping algorithm to determine the cumulative deformation and precise height of each PS point at different time points. This step generates the height information and cumulative settlement information of the initial PS points. The geographic coordinates, cumulative deformation information and height information of each initial PS point are summarized to generate InSAR deformation monitoring point cloud data. The point cloud data contains the precise spatial position, settlement status and height of each PS point in the monitoring area, and is the key input data for subsequent building settlement monitoring and safety risk analysis.

[0118] The interferogram sequence is generated by image pairing, and the interferogram is subjected to phase consistency analysis and denoising to accurately identify the initial PS point in the monitoring area. The application of the phase unwrapping algorithm enables the cumulative deformation and height information of the PS point to be obtained between multi-phase images, achieving high-precision deformation measurement. The InSAR deformation monitoring point cloud data finally generated provides accurate and reliable data support for subsequent building settlement and safety risk assessment. This method ensures the high spatial resolution and long-term deformation monitoring capability of the data, effectively improving the accuracy and stability of building deformation analysis and risk identification.

[0119] This embodiment is based on the permanent scatterer synthetic aperture radar interferometry technology to solve the SAR image of the monitoring area and generate InSAR deformation monitoring point cloud data; the building vector frame data of the monitoring area is unified with the projection coordinate system of the InSAR deformation monitoring point cloud data, and the candidate PS points are determined by setting a buffer zone at the edge of the vector frame and the vector frame range and the measurement point height screening method; if the number of candidate PS points is not less than the preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuations removed; based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established according to the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model, and a time series of the differential settlement index is generated; characteristic indicators are extracted from the time series of the differential settlement index, and the safety risk classification of the buildings in the monitoring area is performed based on the characteristic indicators; wherein the characteristic indicators include the maximum value of the time series, the average change rate, and the over-limit time ratio. This embodiment improves the accuracy of building safety risk classification by utilizing InSAR deformation monitoring point cloud data, accurate building vector frame and PS point screening, stripping seasonal fluctuations, establishing a building differential settlement model, and extracting key characteristic indicators; InSAR technology provides high-precision long-term settlement data, building vector frame and PS point screening ensure the pertinence of the data, stripping seasonal fluctuations removes short-term environmental impacts, and the differential settlement model accurately reflects the settlement characteristics of the building in the X and Y directions. Finally, classification is performed based on multi-dimensional characteristic indicators such as the maximum value of the time series, the average change rate, and the excess time ratio, which comprehensively reflects the building settlement risk characteristics, thereby improving the accuracy of building safety risk classification.

[0120] Based on the above second embodiment, a third embodiment of the building safety risk classification method of the present application is proposed. Figure 3 , Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the building safety risk classification method of the present application.

[0121] In this embodiment, step S2 includes:

[0122] S21: converting the coordinate system of the building vector frame data into the projection coordinate system of the InSAR deformation monitoring point cloud data;

[0123] S22: setting a buffer zone of a preset distance at the boundary of each building vector frame, and performing spatial screening on the initial PS points in the InSAR deformation monitoring point cloud data based on the building vector frame range and the buffer zone to obtain spatially screened PS points;

[0124] S23: Compare the estimated elevation of the spatial screening PS point with the building height range, and based on the comparison result, perform height screening on the spatial screening PS point to determine the candidate PS point.

[0125] It should be noted that the building vector frame data is spatial data that represents the building boundary in vector form, usually including the building's outline and corner coordinates, and is used to locate the building in geographic space. The projection coordinate system is a method of mapping the three-dimensional spatial position on the earth's surface to a two-dimensional plane, which is suitable for accurate position matching of regional geographic data. Different data may use different projection coordinate systems, so coordinate system conversion is required to ensure data consistency. The buffer zone refers to an area set at a certain distance outside the building vector frame boundary, which is used to cover PS points near the edge of the building to increase the number of valid measurement points included. The initial PS point is a permanent scatterer (PS) point with high coherence and reliability in the InSAR deformation monitoring point cloud data, reflecting the stable deformation characteristics of the surface. Elevation refers to the vertical height of a point in space, usually expressed as altitude or relative to a reference plane. In this step, the elevation is used to screen PS points that match the building height. Candidate PS points refer to PS points directly related to the building structure determined through spatial screening and height screening, which are used to further analyze the building settlement and deformation.

[0126] Specifically, the coordinate system of the building vector frame data is converted to the projection coordinate system used by the InSAR deformation monitoring point cloud data to ensure that the two are accurately matched in space. After the coordinate system is unified, a preset buffer zone (e.g., 2 meters or 5 meters) is set outside the building boundary based on each building vector frame to include stable points at the edge of the building. The buffer zone expands the spatial range of the building, making the included PS points more representative, which helps to improve the integrity of the data.

[0127] Furthermore, after unifying the coordinate system and setting the buffer zone, the initial PS points in the InSAR deformation monitoring point cloud data are spatially screened, and the PS points within the building vector frame and its buffer zone are retained to screen out the spatially screened PS points related to the building. Next, the estimated elevations of these spatially screened PS points are compared with the height range of the building, and the PS points that are not within the building height range are screened out to ensure that the selected PS points accurately reflect the settlement characteristics of the building. Through this process, candidate PS points that match the height of the building structure are finally obtained.

[0128] The spatial consistency of the building vector frame data and InSAR point cloud data is ensured through coordinate system conversion, and the buffer setting covers the stable points at the edge of the building, making the screened PS point data more complete. The spatial screening and height screening steps further eliminate PS points that are not related to the building to ensure that the candidate PS points accurately represent the settlement of the building. This precise PS point screening method improves the pertinence of the data and the reliability of the analysis, provides high-quality basic data for subsequent building settlement analysis and safety risk assessment, and improves the monitoring accuracy and effectiveness of risk identification.

[0129] Based on the above second embodiment, in this embodiment, before step S3, the following is further included:

[0130] S3a: comparing the number of the candidate PS points with the preset number threshold;

[0131] S3b: if the number of the candidate PS points is less than the preset number threshold, marking the corresponding building as an uncomputable building;

[0132] S3c: Obtain the maximum cumulative sedimentation value, long-term sedimentation rate and short-term sedimentation rate of the candidate PS point as auxiliary data for manual analysis.

[0133] It should be noted that the quantity threshold refers to the minimum number of data points used to evaluate the effectiveness of building monitoring. If the number of candidate PS points is lower than this threshold, it is considered that the data is insufficient for automatic calculation and analysis. Uncalculatable buildings refer to buildings whose number of candidate PS points does not reach the preset threshold. The data is insufficient to support accurate automatic calculation, so manual analysis is required to assist in judgment. The maximum cumulative settlement value refers to the maximum total settlement amount of the candidate PS point during the monitoring period, reflecting the extreme settlement state of the building. The long-term settlement rate refers to the average rate of change of settlement of the building during the monitoring period, representing the long-term settlement trend of the building. The short-term settlement rate refers to the rate of change of building settlement calculated in a shorter period of time, reflecting the fluctuation of the recent settlement trend of the building.

[0134] Specifically, the number of candidate PS points is compared with the preset number threshold. If the number of candidate PS points reaches or exceeds the threshold, it means that the data volume is sufficient and the automatic calculation and analysis can continue; if the number is lower than the threshold, it is considered that the monitoring data of the building is insufficient and the accuracy of the automatic calculation cannot be ensured. Therefore, the building is marked as an uncomputable building and other analysis methods are used for processing.

[0135] Furthermore, for buildings marked as uncalculated, the key deformation features of candidate PS points are extracted, including the maximum cumulative settlement, long-term settlement rate, and short-term settlement rate. The maximum cumulative settlement is used to quantify the maximum settlement risk of the building; the long-term settlement rate reflects the overall settlement trend of the building; and the short-term settlement rate provides the deformation dynamics of the building in a short period of time. These characteristic data serve as auxiliary information for manual analysis and provide a basis for further risk judgment.

[0136] By comparing the number of candidate PS points with the preset threshold, the basic requirements of data quality are ensured and the accuracy of automatic analysis is improved. Buildings with insufficient candidate points are marked and transferred to manual analysis to avoid calculation errors caused by insufficient data. At the same time, key indicators such as the maximum cumulative settlement, long-term and short-term settlement rates are extracted to provide detailed auxiliary data for manual analysis, so that even when automatic calculation is not feasible, reasonable safety risk assessment can still be carried out. This method ensures the integrity of the building safety monitoring process and the reliability of the analysis results.

[0137] Based on the above second embodiment, in this embodiment, step S5 includes:

[0138] S51: performing data cleaning and data completion on the time series of the differential sedimentation index to obtain a target sequence;

[0139] S52: Obtaining the maximum absolute value of the differential sedimentation index in the target sequence as the maximum value of the time series;

[0140] S53: determining an average change rate according to the total time span of the target sequence and the total change amount of the differential sedimentation index;

[0141] S54: determining the total time of exceeding the limit according to the preset risk threshold, and taking the ratio of the total time of exceeding the limit to the total time of the target sequence as the time ratio of exceeding the limit;

[0142] S55: Compare the maximum value of the time series with a preset maximum value threshold, the average change rate with a preset rate threshold, and the over-limit time ratio with a preset over-limit ratio threshold, and determine the risk classification based on the comparison results.

[0143] Step S55 includes:

[0144] S551: Compare the maximum value of the time series with the preset maximum value threshold, and if the maximum value of the time series is greater than the preset maximum value threshold, mark the maximum value of the time series as a risk indicator;

[0145] S552: Compare the average change rate with the preset rate threshold, and if the average change rate is greater than the preset rate threshold, mark the average change rate as the risk indicator;

[0146] S553: ​​Compare the over-limit time ratio with the preset over-limit ratio threshold, and if the over-limit time ratio is greater than the preset over-limit ratio threshold, mark the over-limit time ratio as the risk indicator;

[0147] S554: Determine risk classification based on the number of risk indicators.

[0148] It should be noted that the differential settlement index time series is a time series data reflecting the uneven settlement characteristics of the building in the X and Y directions calculated based on the building differential settlement model, which is used to analyze the dynamic process of building settlement changes. The maximum value of the time series refers to the maximum absolute value of the differential settlement index recorded in the target sequence, indicating the most serious uneven settlement state of the building during the monitoring period.

[0149] Specifically, the time series of the differential sedimentation index is cleaned to remove outliers and noise, such as short-term large abnormal fluctuations; then the missing values ​​are filled through interpolation or regression models to ensure the integrity and continuity of the time series and generate the target sequence.

[0150] Furthermore, key characteristic indicators are extracted from the target sequence: Maximum value of time series: traverse the target sequence, extract the maximum value of the differential settlement index, record the corresponding time point, and represent the extremely uneven settlement state during the building monitoring period; Average change rate: according to the total time span of the target sequence and the total change of settlement, use the formula average change rate = total change / total time span to calculate the long-term settlement rate of the building; Exceeding time ratio: set the settlement risk threshold, count the number of time points in the target sequence where the differential settlement index exceeds the threshold, and calculate the ratio of the total exceeding time to the total number of time points. The maximum value, average change rate and exceeding time ratio of the time series are compared with the preset thresholds respectively. If a characteristic indicator exceeds the threshold, it is marked as a risk indicator. According to the number and severity of risk indicators, the risk level of the building is comprehensively assessed. The risk level is usually divided into: low risk: all indicators are below the threshold; medium risk: one or more indicators are close to or slightly above the threshold; high risk: multiple indicators obviously exceed the threshold.

[0151] By cleaning and completing the differential settlement index time series, the integrity and reliability of the data are ensured; by extracting key characteristic indicators such as the maximum value of the time series, the average rate of change, and the proportion of over-limit time, the extreme state, long-term trend, and risk exposure time of building settlement are fully reflected. By comparing these indicators with the preset thresholds and combining them with classification rules, building risks can be scientifically classified. This method comprehensively considers the settlement characteristics of different dimensions, which not only improves the accuracy of risk judgment, but also provides a clear basis for building safety management.

[0152] This embodiment is based on the permanent scatterer synthetic aperture radar interferometry technology to solve the SAR image of the monitoring area and generate InSAR deformation monitoring point cloud data; the building vector frame data of the monitoring area is unified with the projection coordinate system of the InSAR deformation monitoring point cloud data, and the candidate PS points are determined by setting a buffer zone at the edge of the vector frame and the vector frame range and the measurement point height screening method; if the number of candidate PS points is not less than the preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuations removed; based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established according to the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model, and a time series of the differential settlement index is generated; characteristic indicators are extracted from the time series of the differential settlement index, and the safety risk classification of the buildings in the monitoring area is performed based on the characteristic indicators; wherein the characteristic indicators include the maximum value of the time series, the average change rate, and the over-limit time ratio. This embodiment improves the accuracy of building safety risk classification by utilizing InSAR deformation monitoring point cloud data, accurate building vector frame and PS point screening, stripping seasonal fluctuations, establishing a building differential settlement model, and extracting key characteristic indicators; InSAR technology provides high-precision long-term settlement data, building vector frame and PS point screening ensure the pertinence of the data, stripping seasonal fluctuations removes short-term environmental impacts, and the differential settlement model accurately reflects the settlement characteristics of the building in the X and Y directions. Finally, classification is performed based on multi-dimensional characteristic indicators such as the maximum value of the time series, the average change rate, and the excess time ratio, which comprehensively reflects the building settlement risk characteristics, thereby improving the accuracy of building safety risk classification.

[0153] For example, in order to help understand the technical concept or technical principle of the building safety risk classification method of the above embodiment, please refer to Figure 4 and Figure 5 , Figure 4 This is a schematic diagram of a local building coordinate system in an embodiment of the building safety risk classification method of the present application; Figure 5This is a schematic diagram of the overall differential settlement model of a building in one embodiment of the building safety risk classification method of the present application; the specific implementation of the building safety risk classification method in one embodiment includes the following steps.

[0154] Step 1: According to the selected monitoring area, SAR images with better time and space baselines are selected for solution. The PS-InSAR permanent scatterer time series interferometry technology is used to generate the interference pattern and complete the estimation of general parameters such as PS point height and accumulated deformation to generate the InSAR deformation monitoring point cloud.

[0155] Step 2: Import the building vector frame data of the target area and unify its projection coordinate system with the InSAR solution data. First, select all PS points within the range by setting a 2m buffer zone at the edge of the vector frame, determine the candidate PS point set located around each building by the vector frame range and the measurement point height screening method, and then classify the PS points into building PS points and non-building PS points according to the estimated elevation of the PS points. The specific judgment formula is as follows:

[0156]

[0157] Among them, H ps ,H threshold They represent the estimated elevation and height threshold of the PS point, respectively. The specific values ​​can be set according to the data situation. To ensure the reliability of the conclusions of subsequent analysis and evaluation, buildings with less than 3 points in the calculation point concentration are recorded as uncalculated buildings, and those with more than 3 points are recorded as calculable buildings.

[0158] Step 3: For incalculable buildings with insufficient number of measuring points, risk classification is performed by manual analysis by analyzing the maximum cumulative settlement value, long-term and short-term settlement rate, etc. of the measuring points on the building, combined with other effective monitoring data. For calculable buildings with more than 3 effective measuring points, the seasonal fluctuation component in the cumulative deformation after solution is stripped, and the nonlinear trend of the measuring points after stripping can be expressed as follows:

[0159]

[0160] in Respectively represent the i-th measuring point t of the building j The periodic fluctuation deformation components extracted by time fitting, a, b, c, d represent the estimated coefficients calculated by nonlinear least squares fitting, t j It indicates the number of days from the current calculation time to the first day of the year, and T indicates the number of days in a whole cycle.

[0161] Step 4: Construct the local coordinate system of the building. The coordinates before and after the conversion are shown as follows: Figure 4As shown. Take two points from the long axis contour line of the building vector frame, calculate the azimuth of the line segment connecting the two points, and further convert it into the angle θ between the long axis of the building and the WE axis through its relative relationship with the coordinate axis of the longitude and latitude coordinate system. Take any PS point in the building measurement point set as the reference origin O. With the reference point as the center, transform the WE axis and NS axis into the XY axis after rotating by θ. The coordinate units of the three axes of this coordinate system are all meters. Take any point P in the PS point set as the reference origin O. i Geographic coordinates (x i ,y i ,h i ) as an example. Its horizontal distance OP from the reference point O is i It can be calculated using Vincenty's formula. i Angle with WE axis It can be calculated by the inverse tangent function of the ratio of its projection distance on the NS axis to its projection distance on the WE axis. i The new coordinates of the point and the relative settlement deformation of the point can be expressed as:

[0162]

[0163] Step 5: Calculate any point P in the measurement point set i (x i ,y i ,h i ) on the XOZ axis. Figure 5 As shown in the figure, V is P i At the foot of the Z axis, when the building is perpendicular to the X axis, x ° differential sedimentation, can be simplified to a triangle OVP i Rotate alpha x °Transformed into triangle OV'P i '. From the overall differential settlement geometric relationship, we know that ∠VOV',∠P i OP i ',∠SV'P i 'All are α x °, then P i Point P is deformed to P due to the overall differential settlement of the building i ', the settlement caused in the X direction can be expressed as:

[0164]

[0165] At the same time, α is superimposed in the Y direction y °Differential sedimentation, P i The total settlement generated by the point can be expressed as:

[0166]

[0167] Through time-series InSAR monitoring, the time series of all PS points of the building have been obtained after removing seasonal fluctuations. j , the estimated parameters of the overall differential settlement of the building represented by all PS points are and constant According to the least squares estimate:

[0168]

[0169] Step 6: Classification of building differential settlement and deformation risks

[0170] Combine the X and Y direction differential settlement indices estimated in the fifth step to obtain the XY comprehensive differential settlement index of the building. Its j The overall settlement index at a given moment is expressed as follows:

[0171]

[0172] S max ,S rate ,S Δt The maximum value, change rate and over-limit time ratio of the time series are the three characteristic indicators of the series:

[0173]

[0174] The present application also provides a building safety risk classification device, please refer to Figure 6 , Figure 6 This is a schematic diagram of the module structure of the building safety risk classification device according to an embodiment of the present application, wherein the building safety risk classification device comprises:

[0175] The image solving module 601 is used to solve the SAR image of the monitoring area based on the permanent scatterer synthetic aperture radar interferometry technology to generate InSAR deformation monitoring point cloud data;

[0176] The candidate point confirmation module 602 is used to unify the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determine the candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method;

[0177] The fluctuation stripping module 603 is used to strip the seasonal fluctuation component in the deformation data of the candidate PS points if the number of the candidate PS points is not less than a preset number threshold, so as to obtain a nonlinear deformation trend with seasonal fluctuation removed;

[0178] The differential settlement module 604 is used to construct a local building coordinate system based on the nonlinear deformation trend with seasonal fluctuations removed, and establish a building differential settlement model according to the local building coordinate system, determine the differential settlement index of the building in the X and Y directions through the building differential settlement model, and generate a time series of the differential settlement index;

[0179] The target module 605 is used to extract characteristic indicators from the time series of the differential settlement index, and classify the safety risks of the buildings in the monitoring area based on the characteristic indicators; wherein the characteristic indicators include the maximum value of the time series, the average rate of change and the proportion of over-limit time.

[0180] The building safety risk classification device provided in the embodiment of the present application adopts the building safety risk classification method in the above embodiment, which can solve the technical problem of how to improve the accuracy of building safety risk classification. Compared with the prior art, the beneficial effects of the building safety risk classification device provided in the embodiment of the present application are the same as the beneficial effects of the building safety risk classification method provided in the above embodiment, and the other technical features in the building safety risk classification device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0181] The present application provides a building safety risk classification device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the building safety risk classification method in the above-mentioned embodiment.

[0182] Reference below Figure 7 , Figure 7 This is a schematic diagram of the device structure of the hardware operating environment involved in the building safety risk classification method in the embodiment of the present application, which shows a schematic diagram of the structure of the building safety risk classification device suitable for implementing the embodiment of the present application. Figure 7 The building safety risk classification device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0183] like Figure 7As shown, the building safety risk classification device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the building safety risk classification device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the building safety risk classification device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a building safety risk classification device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have instead.

[0184] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0185] The building safety risk classification device provided by the present application adopts the building safety risk classification method in the above embodiment, which can solve the technical problem of how to improve the accuracy of building safety risk classification. Compared with the prior art, the beneficial effects of the building safety risk classification device provided by the present application are the same as the beneficial effects of the building safety risk classification method provided by the above embodiment, and the other technical features in the building safety risk classification device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0186] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the building safety risk classification method in the above-mentioned embodiment.

[0187] The above-mentioned computer-readable storage medium may be included in the building safety risk classification device; or it may exist independently without being assembled into the building safety risk classification device.

[0188] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned building safety risk classification method, and can solve the technical problem of how to improve the accuracy of building safety risk classification. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the building safety risk classification method provided in the above-mentioned embodiment, and will not be elaborated here.

[0189] An embodiment of the present application provides a computer program product, including a computer program, which implements the steps of the above-mentioned building safety risk classification method when executed by a processor.

[0190] The computer program product provided in this application can solve the technical problem of how to improve the accuracy of building safety risk classification. Compared with the prior art, the beneficial effects of the computer program product provided in the embodiment of this application are the same as the beneficial effects of the building safety risk classification method provided in the above embodiment, which will not be repeated here.

[0191] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A building safety risk classification method, characterized in that: The method comprises: Based on the permanent scatterer synthetic aperture radar interferometry technology, the SAR image of the monitoring area is solved to generate InSAR deformation monitoring point cloud data; Unify the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determine the candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method; If the number of the candidate PS points is not less than a preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend with seasonal fluctuation removed; Based on the nonlinear deformation trend with seasonal fluctuations removed, a local coordinate system of the building is constructed, and a building differential settlement model is established according to the local coordinate system of the building, and the differential settlement index of the building in the X and Y directions is determined by the building differential settlement model, and a time series of the differential settlement index is generated; Characteristic indicators are extracted from the time series of the differential settlement index, and based on the characteristic indicators, the buildings in the monitoring area are classified for safety risks; wherein the characteristic indicators include the maximum value of the time series, the average rate of change, and the proportion of over-limit time.

2. The method according to claim 1, characterized in that Before the step of solving the SAR image of the monitoring area based on the permanent scatterer synthetic aperture radar interferometry technology to generate InSAR deformation monitoring point cloud data, the method further includes: Based on the location data of the monitoring area, an initial SAR image is acquired from a preset satellite remote sensing data platform; Based on a preset time baseline and a preset space baseline, the initial SAR image is screened to obtain a screened SAR image; The filtered SAR image is preprocessed to obtain the SAR image; wherein the preprocessing includes one or more of radiation correction, geometric correction and registration.

3. The method according to claim 1, characterized in that The step of solving the SAR image of the monitoring area based on the permanent scatterer synthetic aperture radar interferometry technology to generate InSAR deformation monitoring point cloud data includes: Based on the permanent scatterer synthetic aperture radar interferometry technology, the SAR images are paired to generate an interferogram sequence; Determining an initial PS point according to the phase consistency and scattering intensity of the interference patterns in the interference pattern sequence; De-noising the phase time series of the initial PS point to obtain interference phases at multiple time points; Based on the phase unwrapping algorithm, the interference phases of the multiple time points are continuously solved to determine the accumulated deformation information and height information of the initial PS point; The geographic coordinate information of the initial PS point, the accumulated deformation information and the height information are aggregated to generate the InSAR deformation monitoring point cloud data.

4. The method according to claim 3, characterized in that The step of unifying the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determining the candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method, comprises: Converting the coordinate system of the building vector frame data into the projection coordinate system of the InSAR deformation monitoring point cloud data; A buffer zone of a preset distance is set at the boundary of each building vector frame, and based on the building vector frame range and the buffer zone, the initial PS points in the InSAR deformation monitoring point cloud data are spatially screened to obtain spatially screened PS points; The estimated elevation of the spatial screening PS point is compared with the building height range, and based on the comparison result, the spatial screening PS point is height screened to determine the candidate PS point.

5. The method according to claim 1, characterized in that Before the step of stripping seasonal fluctuation components from the deformation data of the candidate PS points to obtain a nonlinear deformation trend without seasonal fluctuations if the number of the candidate PS points is not less than a preset number threshold, the method further includes: Comparing the number of the candidate PS points with the preset number threshold; If the number of the candidate PS points is less than the preset number threshold, the corresponding building is marked as an uncomputable building; The maximum cumulative sedimentation value, long-term sedimentation rate and short-term sedimentation rate of the candidate PS point are obtained as auxiliary data for manual analysis.

6. The method according to claim 1, characterized in that The step of extracting characteristic indicators from the time series of the differential settlement index and classifying the safety risks of the buildings in the monitoring area based on the characteristic indicators comprises: Performing data cleaning and data completion on the time series of the differential sedimentation index to obtain a target sequence; Obtaining the maximum absolute value of the differential sedimentation index in the target sequence as the maximum value of the time series; Determining an average rate of change according to the total time span of the target sequence and the total change of the differential sedimentation index; According to the preset risk threshold, the total time of exceeding the limit is determined, and the ratio of the total time of exceeding the limit to the total time of the target sequence is used as the time ratio of exceeding the limit; The maximum value of the time series is compared with a preset maximum value threshold, the average change rate is compared with a preset rate threshold, and the over-limit time ratio is compared with a preset over-limit ratio threshold, and the risk classification is determined based on the comparison results.

7. The method according to claim 6, characterized in that The step of comparing the maximum value of the time series with a preset maximum value threshold, the average change rate with a preset rate threshold, and the over-limit time ratio with a preset over-limit ratio threshold, and determining the risk classification according to the comparison results includes: Comparing the maximum value of the time series with the preset maximum value threshold, if the maximum value of the time series is greater than the preset maximum value threshold, marking the maximum value of the time series as a risk indicator; Comparing the average change rate with the preset rate threshold, and if the average change rate is greater than the preset rate threshold, marking the average change rate as the risk indicator; Comparing the over-limit time ratio with the preset over-limit ratio threshold, and if the over-limit time ratio is greater than the preset over-limit ratio threshold, marking the over-limit time ratio as the risk indicator; Based on the number of risk indicators, a risk classification is determined.

8. The method according to claim 1, characterized in that If the number of the candidate PS points is not less than a preset number threshold, the seasonal fluctuation component in the deformation data of the candidate PS points is stripped to obtain a nonlinear deformation trend that removes the seasonal fluctuation, comprising: If the number of the candidate PS points is not less than a preset number threshold, extracting deformation data of the candidate PS points; Based on the deformation data, a time series is generated, and the time series is aligned; Based on a preset time series analysis algorithm, detecting the seasonal fluctuation component in the time series; Based on the fluctuation period of the seasonal fluctuation component, a fitting algorithm is used to remove the seasonal fluctuation component from the time series to obtain the nonlinear deformation trend of removing the seasonal fluctuation.

9. A building safety risk classification device, characterized in that: The device comprises: The image solving module is used to solve the SAR image of the monitoring area based on the permanent scatterer synthetic aperture radar interferometry technology to generate InSAR deformation monitoring point cloud data; A candidate point confirmation module is used to unify the building vector frame data of the monitoring area with the projection coordinate system of the InSAR deformation monitoring point cloud data, and determine the candidate PS points by setting a buffer zone at the edge of the vector frame and a measuring point height screening method; A fluctuation stripping module, for stripping seasonal fluctuation components from the deformation data of the candidate PS points if the number of the candidate PS points is not less than a preset number threshold, to obtain a nonlinear deformation trend with seasonal fluctuations removed; A differential settlement module is used to construct a local building coordinate system based on the nonlinear deformation trend with seasonal fluctuations removed, and to establish a building differential settlement model according to the local building coordinate system, to determine the differential settlement index of the building in the X and Y directions through the building differential settlement model, and to generate a time series of the differential settlement index; The target module is used to extract characteristic indicators from the time series of the differential settlement index, and classify the safety risks of the buildings in the monitoring area based on the characteristic indicators; wherein the characteristic indicators include the maximum value of the time series, the average change rate and the proportion of over-limit time.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the building safety risk classification method according to any one of claims 1 to 8 are implemented.

Images