A method, apparatus, system, and storage medium for assessing building stability.

Abstract

This invention provides a method, apparatus, system, and storage medium for building stability assessment, belonging to the field of building monitoring technology. The method includes: performing differential interferometry on raw SAR images to obtain differential interferograms; filtering and analyzing the differential interferograms to obtain high coherence points and initial phase time series; analyzing the deformation phase of the initial phase time series, terrain information, and orbital information to obtain deformation time series and initial deformation rate; importing temperature time series, fitting and removing thermal deformation rates; importing building outline data, filtering building deformation points, constructing and evaluating stability indices for multiple buildings, and finally analyzing the stability indices to obtain multiple building deformation patterns. This invention can provide a more comprehensive stability assessment for buildings, providing important technical support and theoretical basis for urban building stability assessment and structural safety assessment.

Description

Technical Field

[0001] This invention relates to the field of building monitoring technology, specifically to a building stability assessment method, device, system, and storage medium. Background Technology

[0002] Currently, InSAR technology is widely used for monitoring surface deformation in urban areas, providing surface subsidence rates and time-series data with high spatiotemporal resolution. Because it can monitor surface deformation over large areas, InSAR technology can also be applied to the stability analysis of urban buildings. However, building deformation caused by ground subsidence and structural damage often exhibits complex characteristics, potentially including tilting or localized twisting. These complex deformation characteristics make it difficult to directly apply traditional InSAR technology to building stability assessments.

[0003] Deformation measured by InSAR technology is typically along the line-of-sight (LOS) of the satellite observation of the ground. This means that the measured deformation is a projection of the building's actual deformation onto the satellite's line-of-sight, making it difficult to reflect the building's actual settlement, local tilt, torsion, and other specific conditions. Building tilt (e.g., local tilt caused by uneven settlement) and torsion (deformation of the building structure in different directions) are difficult to distinguish directly from LOS measurements, posing a challenge to building stability assessment. Furthermore, the properties of building materials and environmental factors (such as temperature and humidity changes) can also generate spurious deformation signals, which can overlap with the building's actual deformation, further affecting the reliability of using deformation data to assess building stability. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, device, system and storage medium for assessing building stability, in order to address the shortcomings of the prior art.

[0005] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A method for assessing building stability, comprising the following steps:

[0006] Import multiple raw SAR images, perform differential interferometry on all the raw SAR images, and obtain multiple differential interferograms;

[0007] By screening and analyzing all the differential interferograms, multiple high coherence points and the initial phase time series corresponding to each high coherence point are obtained;

[0008] Import multiple terrain information and multiple orbit information corresponding to each of the high coherence points, and analyze the deformation phase of the initial phase time series corresponding to each of the high coherence points, the multiple terrain information corresponding to each of the high coherence points, and the multiple orbit information corresponding to each of the high coherence points to obtain the deformation time series and the initial deformation rate corresponding to each of the high coherence points.

[0009] Import multiple temperature data corresponding to each of the high coherence points, and perform rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and the multiple temperature data corresponding to each of the high coherence points to obtain the thermal deformation rate corresponding to each of the high coherence points.

[0010] The difference between the initial deformation rate corresponding to each of the high coherence points and the thermal deformation rate corresponding to each of the high coherence points is calculated to obtain the initial net deformation rate corresponding to each of the high coherence points.

[0011] Import the projection ratio corresponding to each of the high coherence points, and perform vertical decomposition on the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points to obtain the vertical net settlement rate corresponding to each of the high coherence points.

[0012] Import building outline vector data, and based on the building outline vector data and all the vertical net settlement rates, filter out multiple target building settlement points and the target net settlement rate corresponding to each of the target building settlement points from all the high coherence points.

[0013] A stability analysis of the building is performed on all the target building settlement points and all the target net settlement rates, and the analysis results are used as the evaluation results of building stability.

[0014] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: A building stability assessment device, comprising:

[0015] The import module is used to import multiple raw SAR images;

[0016] The differential interferometry module is used to perform differential interferometry on all the original SAR images to obtain multiple differential interferograms;

[0017] The screening and analysis module is used to screen and analyze all the differential interferograms to obtain multiple high coherence points and the initial phase time series corresponding to each of the high coherence points;

[0018] The import module is also used to import multiple terrain information corresponding to each of the high coherence points and multiple track information corresponding to each of the high coherence points;

[0019] The deformation phase analysis module is used to analyze the initial phase time series corresponding to each of the high coherence points, multiple terrain information corresponding to each of the high coherence points, and multiple orbital information corresponding to each of the high coherence points, respectively, to obtain the deformation time series corresponding to each of the high coherence points and the initial deformation rate corresponding to each of the high coherence points.

[0020] The import module is also used to import multiple temperature data corresponding to each of the high coherence points;

[0021] The rate analysis module is used to perform rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and multiple temperature data corresponding to each of the high coherence points, to obtain the thermal deformation rate corresponding to each of the high coherence points.

[0022] The difference calculation module is used to calculate the difference between the initial deformation rate corresponding to each of the high coherence points and the thermal deformation rate corresponding to each of the high coherence points, so as to obtain the initial net deformation rate corresponding to each of the high coherence points.

[0023] The import module is also used to import the projection ratio values ​​corresponding to each of the highly coherent points;

[0024] The vertical decomposition module is used to perform vertical decomposition on the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points, so as to obtain the vertical net settlement rate corresponding to each of the high coherence points.

[0025] The import module is also used to import building outline vector data;

[0026] The filtering module is used to filter out multiple target building settlement points and the target net settlement rate corresponding to each target building settlement point from all the high coherence points based on the building outline vector data and all the vertical net settlement rates.

[0027] The evaluation result acquisition module is used to perform stability analysis on all the target building settlement points and all the target net settlement rates, and use the analysis results as the evaluation results of building stability.

[0028] Based on the above-mentioned building stability assessment method, the present invention also provides a building stability assessment system.

[0029] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: a building stability assessment system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the building stability assessment method as described above.

[0030] Based on the above-mentioned building stability assessment method, the present invention also provides a computer-readable storage medium.

[0031] Another technical solution of the present invention to solve the above-mentioned technical problems is as follows: a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the building stability assessment method as described above.

[0032] The beneficial effects of this invention are as follows: Differential interferometry is obtained by differentially interferometric analysis of the original SAR image; high coherence points and initial phase time series are obtained through screening and analysis of the differential interferometry; deformation time series and initial deformation rate are obtained through deformation phase analysis of the initial phase time series, terrain information, and orbital information; thermal deformation rate is obtained through rate analysis of the deformation time series, initial deformation rate, and temperature data; the initial net deformation rate is calculated by the difference between the initial deformation rate and the thermal deformation rate; vertical net settlement rate is obtained by vertical decomposition of the initial net deformation rate and projection ratio corresponding to each of the high coherence points; and the vertical net settlement rate is obtained based on the building outline vector data and the vertical net settlement rate. By selecting target building settlement points and target net settlement rates from high coherence points, and conducting building stability analysis on these points and rates, the results are used as the assessment of building stability. This approach enables the interpretation of multi-level building settlement patterns from the overall to the local perspective, assessing the potential risk distribution within the building, its overall stability, and local risks. It provides a more comprehensive stability assessment for buildings, offering crucial technical support and theoretical basis for urban building stability and structural safety assessments. This significantly improves the accuracy and reliability of building stability monitoring in practical applications, demonstrating broad practical application value and promising prospects for wider adoption. Attached Figure Description

[0033] Figure 1 A schematic flowchart of the building stability assessment method provided in an embodiment of the present invention;

[0034] Figure 2 This is a block diagram of a building stability assessment device provided in an embodiment of the present invention. Detailed Implementation

[0035] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0036] Figure 1 This is a flowchart illustrating a building stability assessment method provided in an embodiment of the present invention.

[0037] like Figure 1 As shown, a building stability assessment method includes the following steps:

[0038] Import multiple raw SAR images, perform differential interferometry on all the raw SAR images, and obtain multiple differential interferograms;

[0039] By screening and analyzing all the differential interferograms, multiple high coherence points and the initial phase time series corresponding to each high coherence point are obtained;

[0040] Import multiple terrain information and multiple orbit information corresponding to each of the high coherence points, and analyze the deformation phase of the initial phase time series corresponding to each of the high coherence points, the multiple terrain information corresponding to each of the high coherence points, and the multiple orbit information corresponding to each of the high coherence points to obtain the deformation time series and the initial deformation rate corresponding to each of the high coherence points.

[0041] Import multiple temperature data corresponding to each of the high coherence points, and perform rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and the multiple temperature data corresponding to each of the high coherence points to obtain the thermal deformation rate corresponding to each of the high coherence points.

[0042] The difference between the initial deformation rate corresponding to each of the high coherence points and the thermal deformation rate corresponding to each of the high coherence points is calculated to obtain the initial net deformation rate corresponding to each of the high coherence points.

[0043] Import the projection ratio corresponding to each of the high coherence points, and perform vertical decomposition on the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points to obtain the vertical net settlement rate corresponding to each of the high coherence points.

[0044] Import building outline vector data, and based on the building outline vector data and all the vertical net settlement rates, filter out multiple target building settlement points and the target net settlement rate corresponding to each of the target building settlement points from all the high coherence points.

[0045] A stability analysis of the building is performed on all the target building settlement points and all the target net settlement rates, and the analysis results are used as the evaluation results of building stability.

[0046] It should be understood that importing time-series SAR images of the urban area (i.e., the original SAR images) and performing differential interferometry on all time-series SAR images (i.e., the original SAR images) will yield multiple differential interferograms (if there are n SAR images, n-1 differential interferograms will be obtained after differential interferometry).

[0047] Specifically, the thermal deformation component (i.e., thermal deformation rate) is removed from the observed deformation (i.e., initial deformation rate) to obtain the net settlement rate (i.e., initial net deformation rate), as shown in the following formula:

[0048] v struct (t)=v obs (t)-v temp (t).

[0049] In the above embodiments, differential interferograms are obtained by differential interferometry of the original SAR images. High coherence points and initial phase time series are obtained through screening and analysis of the differential interferograms. Deformation phase analysis of the initial phase time series, terrain information, and orbital information yields deformation time series and initial deformation rate. Rate analysis of the deformation time series, initial deformation rate, and temperature data yields thermal deformation rate. The difference between the initial deformation rate and the thermal deformation rate is calculated to obtain the initial net deformation rate. Vertical decomposition of the initial net deformation rate and projection ratio corresponding to each of the high coherence points yields the vertical net settlement rate. Based on the building outline vector data and the vertical net settlement rate, the vertical net settlement rate is calculated from the high... The target building settlement point and target net settlement rate are selected from the coherent points. The building stability is analyzed based on the target building settlement point and target net settlement rate, and the analysis results are used as the building stability assessment results. This realizes the interpretation of multi-level building settlement patterns from the whole to the part, and assesses the potential risk distribution inside the building, the overall stability of the building, and the local risks of the building. It can provide a more comprehensive stability assessment for buildings, and provide important technical support and theoretical basis for urban building stability assessment and structural safety assessment. In practical applications, it greatly improves the accuracy and reliability of building stability monitoring, and has broad practical application value and promotion prospects.

[0050] Optionally, as an embodiment of the present invention, the differential interferogram includes a plurality of initial pixels, an initial amplitude value corresponding to each initial pixel, and an initial phase value corresponding to each initial pixel.

[0051] The process of filtering and analyzing all the differential interferograms to obtain multiple high-coherence points and the initial phase time series corresponding to each high-coherence point includes:

[0052] The standard deviations of multiple initial amplitude values ​​with the same initial pixel correlation in all the differential interferograms are calculated to obtain the amplitude standard deviations with the same initial pixel correlation in all the differential interferograms.

[0053] The average value of multiple initial amplitude values ​​with the same initial pixel correlation in all the differential interferograms is calculated to obtain the average amplitude value with the same initial pixel correlation in all the differential interferograms.

[0054] The amplitude dispersion index of all differential interferograms with the same initial pixel correlation is obtained by calculating the amplitude standard deviation and amplitude mean of all differential interferograms with the same initial pixel correlation using the first formula. The first formula is:

[0055]

[0056] Among them, D A σ is the amplitude dispersion exponent for all differential interferograms with the same initial pixel A correlation. A μ is the amplitude standard deviation of the correlation of all differential interferograms with the same initial pixel A. A The mean amplitude of the correlation of all initial pixel points A in the differential interferograms;

[0057] If the amplitude dispersion index is greater than the preset amplitude threshold, then all the initial pixels corresponding to the amplitude dispersion index are taken as candidate pixels, and the initial phase value corresponding to the candidate pixels is taken as a candidate phase value, thereby obtaining multiple candidate pixels and multiple candidate phase values ​​with the same candidate pixel correlation in all the differential interferograms.

[0058] The temporal coherence coefficients of the same candidate pixel correlation in all the differential interferograms are obtained by calculating the candidate phase values ​​with the same candidate pixel correlation in all the differential interferograms using the second equation. The second equation is:

[0059]

[0060] Where, γ t,B Let N be the temporal coherence coefficient of candidate pixel B with the same correlation in all differential interferograms, and N be the total number of differential interferograms. For the i-th differential interferogram, there are candidate phase values ​​with the same candidate pixel B correlation.

[0061] If the temporal coherence coefficient is greater than the preset coherence coefficient threshold, then all the candidate pixels corresponding to the temporal coherence coefficient are taken as pixels to be processed, and the candidate phase values ​​corresponding to the pixels to be processed are taken as phase values ​​to be processed, thereby obtaining multiple pixels to be processed and multiple phase values ​​to be processed that have the same correlation with the pixels to be processed in all the differential interferograms.

[0062] The probability of each pixel to be processed is calculated by using a pre-constructed probability distribution model to obtain the pixel probability value corresponding to each pixel to be processed.

[0063] If the pixel probability value is greater than a preset probability threshold, then the pixel to be processed corresponding to the pixel probability value is taken as the processed pixel, and the phase value to be processed corresponding to the processed pixel is taken as the target phase value, thereby obtaining multiple processed pixels and the target phase value corresponding to each processed pixel.

[0064] All processed pixels in the differential interferograms are grouped together as high coherence points, thereby obtaining multiple high coherence points and multiple target phase values ​​corresponding to each high coherence point;

[0065] An initial phase time series corresponding to each of the high coherence points is constructed by using multiple target phase values ​​corresponding to each of the high coherence points.

[0066] It should be understood that each differential interferogram includes multiple pixels (i.e., initial pixels), and each pixel (i.e., initial pixel) corresponds to an amplitude value (i.e., initial amplitude value) and a phase value (i.e., initial phase value).

[0067] Specifically, the selection of candidate pixels for screening highly coherent points first relies on amplitude stability analysis. Inputting interferogram data, for each pixel (i.e., the initial pixel), the change in its amplitude value over time is calculated. The amplitude dispersion index (DA), which measures amplitude stability by considering the correlation between pixels with the same target characteristics, is introduced. Its formula is:

[0068]

[0069] Where the numerator is the standard deviation of the amplitude values ​​in the time series of the pixel, and the denominator is the mean of the amplitude values. The smaller this exponent is, the smaller the amplitude variation of the pixel, which means that the pixel is more likely to be a highly coherent point. Therefore, a threshold T is set. a (i.e., preset amplitude threshold), satisfying D a <T a The pixels are set as candidate pixels.

[0070] It should be understood that the temporal phase stability of the selected candidate points (i.e. candidate pixels) is analyzed by the StaMPS algorithm. First, the temporal phase coherence is calculated through temporal phase stability analysis to measure the stability of the interferogram phase.

[0071] Specifically, for each pixel (i.e. the pixel to be processed), the probability of each pixel (i.e. the pixel to be processed) becoming a highly coherent point is quantified by a statistical probability model.

[0072] Specifically, the probability distribution model of the deformation signal and noise signal is pre-constructed to calculate and set a probability threshold (i.e., a preset probability threshold). Pixels with a probability greater than the threshold are retained. Finally, error points are removed by the spatial coherence of adjacent pixels, and pixels (i.e., pixels to be processed) are further screened to obtain stable high coherence points.

[0073] In the above embodiments, all differential interferograms are screened and analyzed to obtain multiple high coherence points and the initial phase time series corresponding to each high coherence point. This enables the interpretation of multi-level building settlement patterns from the overall to the local level, assesses the potential risk distribution inside the building, the overall stability of the building, and the local risks of the building. It can provide a more comprehensive stability assessment for buildings and provides important technical support and theoretical basis for urban building stability assessment and structural safety assessment.

[0074] Optionally, as an embodiment of the present invention, the process of analyzing the deformation phase of the initial phase time series corresponding to each of the high coherence points, the multiple terrain information corresponding to each of the high coherence points, and the multiple orbital information corresponding to each of the high coherence points to obtain the deformation time series corresponding to each of the high coherence points and the initial deformation rate corresponding to each of the high coherence points includes:

[0075] The network flow algorithm is used to perform phase unwrapping processing on the initial phase time series corresponding to each of the high coherence points to obtain the unwrapped phase time series corresponding to each of the high coherence points.

[0076] The unwrapped phase time series corresponding to each of the high coherence points are filtered to obtain the deformation phase series corresponding to each of the high coherence points.

[0077] A linear deformation model is constructed using all the aforementioned deformation phase sequences;

[0078] The linear deformation model is fitted using the least squares algorithm to obtain the deformation time series corresponding to each of the high coherence points and the initial deformation rate corresponding to each of the high coherence points.

[0079] As should be understood, phase unwrapping is a technique for recovering the absolute phase (i.e., eliminating 2π ambiguity) from a measured wrapped phase (typically between [-π, π] or [0, 2π]) that directly reflects the physical properties of interest. This technique plays a crucial role in applications such as Interferometric Synthetic Aperture Radar (InSAR), Magnetic Resonance Imaging (MRI), and Optical Interferometry, and can be considered one of the core technologies in these applications. Especially in the field of InSAR, the quality of the phase unwrapping results directly determines the accuracy of subsequently acquired terrain elevation and deformation information.

[0080] Specifically, after obtaining the high coherence points, the deformation signal is extracted by phase unwrapping and interferometric phase decomposition using the phase time series of the high coherence points (i.e., the initial phase time series) and other auxiliary information (terrain information and orbital information). Atmospheric, orbital and terrain error signals are removed. Then, the deformation time series model is performed on the deformation signal of the high coherence points to obtain the deformation rate (i.e., the initial deformation rate) and the deformation time series.

[0081] In the above embodiments, deformation time series and initial deformation rate are obtained by analyzing the deformation phase of the initial phase time series, terrain information and orbit information, achieving millimeter-level deformation monitoring accuracy, realizing the interpretation of multi-level building settlement patterns from the whole to the local, and assessing the potential risk distribution inside the building, the overall stability of the building and the local risks of the building.

[0082] Optionally, as an embodiment of the present invention, the process of performing rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and multiple temperature data corresponding to each of the high coherence points to obtain the thermal deformation rate corresponding to each of the high coherence points includes:

[0083] By fitting the deformation time series corresponding to each of the high coherence points and multiple temperature data corresponding to each of the high coherence points using a pre-constructed partial least squares regression model, the thermal deformation components corresponding to each of the high coherence points are obtained.

[0084] The thermal deformation rate corresponding to each of the highly coherent points is obtained by taking the derivative of the thermal deformation component.

[0085] Specifically, to accurately eliminate the thermal expansion and contraction deformation effect caused by temperature changes, this invention introduces a partial least squares regression (PLSR) model into the thermal deformation removal process. This model performs dimensionality reduction analysis on the deformation time series and identifies the part most strongly correlated with the temperature time series T(t), thereby extracting the thermal deformation component Δv. temp (t) and thermal deformation rate v temp (t).

[0086] The PLSR model uses the temperature time series T(t) and the deformation time series Δv obs (t) is the input. Principal component analysis is used to reduce the multidimensional deformation time series to the principal component space, from which the deformation component most relevant to temperature change is extracted. Observed deformation Δv obs (t) is expressed as the structural settlement component Δv struct (t) and thermal deformation component Δv temp Superposition of (t):

[0087] Δv obs (t)=Δv struct (t)+Δv temp (t),

[0088] PLSR finds the temperature time series T(t) and the deformation time series Δv obs The principal component correlation between (t) maps the deformation variables to a new principal component space, expressed as:

[0089] Δv PLS (t)=P·Δv obs (t),

[0090] T PLS (t) = Q·T(t),

[0091] Where P and Q are the projection matrices of the deformation and temperature data; Δv PLS (t) and T PLS (t) represent the deformation and temperature in the principal component space, respectively. In the principal component space, PLSR establishes the relationship between deformation and temperature through linear regression, and the fitted equation is:

[0092] Δv PLS (t)=a·T PLS (t)+b,

[0093] Where a and b are fitting coefficients, determined by the least squares method.

[0094] After fitting, the principal component Δv of thermal deformation is transformed by inverse transformation. PLS(t) Transform back to the original data space to obtain the thermal deformation component Δv temp (t), the formula is as follows:

[0095] Δv temp (t)=P -1 ·Δv PLS (t),

[0096] Furthermore, by analyzing the thermal deformation component Δv temp (t) is differentiated over time to calculate the thermal deformation rate, as shown in the following formula:

[0097]

[0098] In the above embodiments, rate analysis of deformation time series, initial deformation rate and temperature data is performed to obtain thermal deformation rate, which accurately separates pseudo-deformation caused by thermal expansion and contraction effect, effectively reduces the interference of temperature change on deformation analysis, ensures the physical authenticity of deformation rate data, and provides high-precision basic data for subsequent quantitative analysis of building deformation characteristics.

[0099] Optionally, as an embodiment of the present invention, the process of vertically decomposing the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points to obtain the vertical net settlement rate corresponding to each of the high coherence points includes:

[0100] By vertically decomposing the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points using the third equation, the vertical net settlement rate corresponding to each of the high coherence points is obtained. The third equation is:

[0101]

[0102] Among them, v vertical,c v is the vertical net settling velocity corresponding to the Cth high coherence point. LOS,c Let cos(θ) be the initial net deformation rate corresponding to the Cth high coherence point. c ) represents the projection ratio corresponding to the Cth highly coherent point.

[0103] It should be understood that, because InSAR deformation rate has projection characteristics along the line-of-sight (LOS), it cannot directly reflect the vertical settlement behavior of a building. Therefore, this invention performs vertical decomposition of the deformation rate at the PS point. Assume the radar incident angle is θ. j Then PS point p j Vertical net settlement rate v vertical,j Expressed as:

[0104]

[0105] Among them, v LOS,j Let cos(θ) be the deformation rate along the line of sight. j The projection ratio is given by the line-of-sight direction and the vertical direction. This formula is used to decompose each deformation rate (i.e., the initial net deformation rate) to generate vertical settlement rate data.

[0106] In the above embodiments, the initial net deformation rate and projection ratio are vertically decomposed to obtain the vertical net settlement rate, which effectively reduces the interference of temperature changes on deformation analysis, ensures the physical authenticity of deformation rate data, and provides high-precision basic data for subsequent quantitative analysis of building deformation characteristics.

[0107] Optionally, as an embodiment of the present invention, the process of selecting multiple target building settlement points and target net settlement rates corresponding to each of the target building settlement points from all the highly coherent points based on the building profile vector data and all the vertical net settlement rates includes:

[0108] S71: Based on the judgment criteria and the building outline vector data, multiple initial building deformation points are selected from all the highly coherent points. The judgment criteria are:

[0109]

[0110] Where, p d Let x be the d-th initial building deformation point. c Let y be the x-axis coordinate of the c-th high coherence point. c Let Γ be the Y-axis coordinates of c highly coherent points, and Γ be the building outline vector data;

[0111] S72: Calculate the distance between each of the initial building deformation points and any remaining initial building deformation point to obtain the distances between multiple initial neighboring points corresponding to each of the initial building deformation points;

[0112] S73: Filter the minimum value of the distances between multiple initial neighboring points corresponding to each of the initial building deformation points to obtain the target neighboring point distances corresponding to each of the initial building deformation points;

[0113] S74: When the target neighbor distance is less than or equal to the preset neighbor distance threshold, the initial building deformation point corresponding to the target neighbor distance is used as the processed building deformation point, thereby obtaining multiple processed building deformation points.

[0114] S75: Count the number of all the processed building deformation points to obtain the total number of processed building deformation points;

[0115] S76: Determine whether the total number of building deformation points after processing is greater than or equal to the preset number of building deformation points. If not, re-import the building outline vector data and return to S71. If yes, use the processed building deformation points as target building settlement points to obtain multiple target building settlement points, and use the vertical net settlement rate corresponding to the target building settlement points as the target net settlement rate to obtain the target net settlement rate corresponding to each target building settlement point.

[0116] It should be understood that the extraction of building settlement points (PS points) involves selecting a set of high-quality PS points related to the building unit from the InSAR results data to form a settlement rate data table for the building unit. This invention extracts PS point data that accurately reflects the building's settlement behavior through steps such as geometric spatial analysis, physical meaning decomposition, and data filtering and optimization, laying the foundation for the analysis of building unit settlement characteristics.

[0117] Specifically, the input consists of building outline vector data B and a set of PS point data P′ (i.e., high coherence points) generated by InSAR, where B = B1, B2, ..., B n This represents n building units, each building unit B i The boundary is represented as a polygon Γ i ;P′=p1,p2,…,p m This represents m points PS, where each point p j Includes geographic coordinates (x j ,y j ), deformation rate v LOS,j coherence γ j and time series Δv obs,j (t). Through geometric spatial analysis, the set P of PS points overlapping with the building unit is selected. i (i.e., multiple initial building deformation points), satisfying the following condition:

[0118]

[0119] That is, PS point (x) j ,y j The geographical location must fall within building unit B. i Boundary Γ i (i.e., building outline vector data).

[0120] Specifically, to improve data quality, this invention performs the following optimization processing on the selected PS point set Pi (i.e., multiple initial building deformation points). Isolated PS points are removed through spatial analysis based on inter-point distances. Point p j The nearest neighbor distance is d min,j (i.e., the distance between the target and its neighboring points), if d min,j (i.e., the distance to the target neighbor) exceeds the preset distance threshold dthreshold (i.e., a preset neighbor distance threshold) will then place point p j These are defined as isolated points and removed from the set. After these two steps, an optimized set of PS points Pi′ (i.e., multiple processed building deformation points) is formed, where each PS point p j Includes geographic location (x j ,y j Vertical settlement rate v vertical,j coherence γ j and time series Δv obs,j (t).

[0121] Specifically, if a certain building unit B i The number of PS points (i.e., the total number of building deformation points after processing) is insufficient, i.e., |Pi′| <N threshold If N is not found, then remove the building unit. threshold This is the preset minimum number of PS points (i.e., the preset number of building deformation points). Furthermore, if the area A of the building unit... i Less than the preset area threshold A threshold This is also removed from the analysis. The set of remaining building units is defined as B′, and its corresponding PS point table is T. i (i.e., multiple target building settlement points).

[0122] Finally, the PS point table T for each building unit i Represented as:

[0123] T i ={(p j ,v vertical,j ,γ j ,Δv obs,j (t))|p j ∈P i ′},

[0124] Among them, T i Summary of Building B i All valid PS point information, including geographical location, vertical settlement rate, coherence, and time series.

[0125] In the above embodiments, multiple target building settlement points and target net settlement rates are selected from high coherence points based on building outline vector data and vertical net settlement rate. This effectively improves the reliability of building unit analysis, eliminates invalid or statistically insufficient buildings, ensures the scientific nature and accuracy of the analysis results, and provides high-quality data support for the quantification of building settlement characteristics and pattern classification.

[0126] Optionally, as an embodiment of the present invention, the process of performing a stability analysis on all the target building settlement points and all the target net settlement rates, and using the analysis results as the assessment results of building stability, includes:

[0127] S81: The building deformation rate distribution index is obtained by calculating the building deformation rate distribution index for all target building settlement points and all target net settlement rates using the fourth formula. The fourth formula is:

[0128]

[0129] Where G is the building deformation rate distribution index, M is the total number of settlement points of the target building, and v f Let v be the target net settlement rate corresponding to the f-th target building settlement point. g Let be the target net settlement rate corresponding to the settlement point of the g-th target building;

[0130] S82: Calculate the average net settlement rate of all the targets to obtain the average deformation rate;

[0131] S83: Determine whether the building deformation rate distribution index is less than or equal to the preset building deformation rate distribution threshold. If yes, proceed to S84; otherwise, proceed to S85.

[0132] S84: Determine whether the average deformation rate is less than or equal to a preset deformation rate threshold. If yes, use the first preset evaluation result as the evaluation result of building stability; otherwise, use the second preset evaluation result as the evaluation result of building stability.

[0133] S85: The fifth formula is used to calculate the horizontal distances to each of the target building settlement points and any remaining target building settlement point, respectively, to obtain multiple horizontal distances corresponding to each of the target building settlement points. The fifth formula is:

[0134]

[0135] Among them, L fg Let x be the horizontal distance between the settlement point of the f-th target building and the settlement point of the g-th target building. f Let x be the X-axis coordinate of the settlement point of the f-th target building. g Let y be the X-axis coordinate of the g-th target building settlement point. f Let y be the Y-axis coordinate of the f-th target building settlement point. g Let g be the Y-axis coordinate of the settlement point of the g-th target building;

[0136] S86: The differential deformation rate is calculated by using the sixth equation to calculate the differential deformation rate corresponding to each of the stated target net settlement rates and any remaining target net settlement rate, respectively. The sixth equation is:

[0137] Δv fg =|v f -v g |,

[0138] Where, Δv fg Let v be the differential deformation rate between the settlement point of the f-th target building and the settlement point of the g-th target building. f Let v be the target net settlement rate corresponding to the f-th target building settlement point. g Let be the target net settlement rate corresponding to the settlement point of the g-th target building;

[0139] S87: Determine whether each horizontal distance is greater than a preset horizontal distance threshold. If yes, proceed to S88; otherwise, proceed to S89.

[0140] S88: The tilt rate corresponding to each horizontal distance is calculated by using the seventh formula to calculate the tilt rate for each horizontal distance and the differential deformation rate corresponding to each horizontal distance. The seventh formula is:

[0141]

[0142] in,

[0143] Where, θ tilt-fg Let θ be the tilt rate between the settlement point of the f-th target building and the settlement point of the g-th target building. fg Let Δv be the angular distortion value between the settlement point of the f-th target building and the settlement point of the g-th target building, t1 be the preset first sampling time, and Δv be the angle distortion value between the settlement points of the f-th and g-th target buildings. fg Let L be the differential deformation rate between the settlement point of the f-th target building and the settlement point of the g-th target building. fg Let f be the horizontal distance between the settlement point of the f-th target building and the settlement point of the g-th target building;

[0144] S89: The torsion rate corresponding to each horizontal distance is calculated by using the eighth formula to calculate the torsion rate for each horizontal distance and the differential deformation rate corresponding to each horizontal distance. The eighth formula is:

[0145]

[0146] in,

[0147] Where, θ warp-fgLet θ be the torsion rate between the settlement point of the f-th target building and the settlement point of the g-th target building. fg Let Δv be the angular distortion value between the settlement point of the f-th target building and the settlement point of the g-th target building, t2 be the preset second sampling time, and Δv be the angle distortion value between the settlement points of the f-th and g-th target buildings. fg Let L be the differential deformation rate between the settlement point of the f-th target building and the settlement point of the g-th target building. fg Let f be the horizontal distance between the settlement point of the f-th target building and the settlement point of the g-th target building;

[0148] S810: Filter out the maximum value among all the tilt rates to obtain the maximum tilt rate;

[0149] S811: Filter out the maximum value among all the stated twist rates to obtain the maximum twist rate;

[0150] S812: When the average deformation rate is less than or equal to the preset deformation rate threshold, and the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, then the third preset evaluation result is taken as the evaluation result of building stability.

[0151] S813: When the average deformation rate is less than or equal to the preset deformation rate threshold, and the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is less than or equal to the preset torsion rate threshold, then the fourth preset evaluation result is taken as the evaluation result of building stability.

[0152] S814: When the average deformation rate is less than or equal to the preset deformation rate threshold, the maximum tilt rate is less than or equal to the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, then the fifth preset evaluation result is taken as the evaluation result of building stability.

[0153] S815: When the average deformation rate is greater than the preset deformation rate threshold, the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, the sixth preset evaluation result shall be used as the evaluation result of building stability.

[0154] S816: When the average deformation rate is greater than the preset deformation rate threshold, the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is less than or equal to the preset torsion rate threshold, the seventh preset evaluation result shall be used as the evaluation result of building stability.

[0155] S817: When the average deformation rate is greater than the preset deformation rate threshold, and the maximum tilt rate is less than or equal to the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, then the eighth preset evaluation result is taken as the evaluation result of building stability.

[0156] Preferably, the preset building settlement rate distribution threshold can be 0.5, the preset settlement rate threshold can be 4, and the preset tilt rate threshold can be 1.0 × 10⁻⁶. -3 The preset torsion rate threshold can be 2.5 × 10⁻⁶. - 3 The preset horizontal distance threshold can be 10 rad / yr.

[0157] It should be understood that by combining the two core indicators, Gini coefficient and angular distortion, the uniform and non-uniform settlement characteristics of a building are quantified, thereby identifying the overall stability of the building and potential risks of abnormal settlement. These indicators can not only describe the overall settlement state of the building, but also provide precise evidence for identifying non-uniform settlement patterns such as local damage, tilting, or torsion.

[0158] Specifically, the building settlement rate distribution index, or Gini coefficient, reflects the uniformity of the settlement rate distribution at the PS points within a building. Assume the set of PS points for building k is v. vertical,k =v1,v2,…,v n , where v i Let represent the vertical settlement rate (i.e., the target net settlement rate) at the i-th PS point. The formula for calculating the Gini coefficient (i.e., the building settlement rate distribution index) is:

[0159]

[0160] Where n is the number of PS points within the building, |v i -v j | represents the absolute value of the speed difference between any two points. This is the sum of the absolute values ​​of the settlement velocities at points PS, used to normalize the results. During calculation, the rate difference matrix D is first constructed. ij =|v i -v j | where i,j∈1,2,…,n. The sum of rate differences is obtained by summing all values ​​in matrix D, as shown in the following equation:

[0161]

[0162] Next, calculate the sum of the absolute values ​​of the rates, as shown in the following formula:

[0163]

[0164] Finally, S D and S V Substituting into the Gini coefficient formula, the settlement rate distribution index G of the building is calculated. The value of the Gini coefficient ranges between [0,1], where G=0 indicates complete homogeneity (all PS points have the same settlement rate), and G=1 indicates complete heterogeneity (the rate of a single PS point accounts for a large proportion of the total).

[0165] After quantifying the uniformity of settlement distribution, this invention introduces the calculation of angular distortion to further analyze the non-uniform settlement characteristics of buildings. Angular distortion is a core indicator characterizing the non-uniform settlement gradient of a building; its physical meaning is the angle formed by the ratio of the vertical settlement rate difference between two points P and PS to the horizontal distance. For point P of PS... i and P j Their settling rates are v i and v j The horizontal distance is L ij Angular distortion θ ij The formula for calculating (i.e., angular distortion value) is:

[0166]

[0167] Wherein, horizontal distance L ij Calculated from the coordinates of the two points:

[0168]

[0169] Angular distortion value θ ij (i.e., angular distortion value) describes the settlement rate gradient between point pairs and is used to identify the overall tilt or local twist characteristics of a building.

[0170] It should be understood that the first preset assessment result is that the building under test is a structurally stable building; the second preset assessment result is that the building under test is a building with overall settlement; the third preset assessment result is that the building under test is a locally anomalous building, and the building under test may be affected by complex foundation conditions or load distribution, with high potential risks; the fourth preset assessment result is that the building under test is a locally anomalous building, and the building under test is mainly affected by non-uniform distribution of foundation bearing capacity or underground activity; the fifth preset assessment result is that the building under test is a locally anomalous building, and the building under test... The sixth preset assessment result indicates that the building under test is a building with severe uneven settlement, and the building under test may be affected by complex foundation conditions or load distribution, with high potential risks. The seventh preset assessment result indicates that the building under test is a building with severe uneven settlement, and the building under test is mainly affected by the non-uniform distribution of foundation bearing capacity or underground activity. The eighth preset assessment result indicates that the building under test is a building with severe uneven settlement, and the building under test may have foundation instability or abnormal structural stress in some areas.

[0171] Specifically, this invention uses the Gini index, average settlement rate, differential settlement rate, and distance between point pairs as core classification criteria to gradually construct a classification system for building settlement patterns, ultimately clarifying the building's settlement type and potential risk distribution. In the first stage of classification, a preliminary classification of settlement patterns is performed based on the overall characteristics of the building. The core of this stage lies in the joint analysis of the Gini index and average settlement rate. For each building B... k Let its Gini exponent be G. k The average settling rate is The classification rules are as follows:

[0172] 1. If G k ≤0.5 and This indicates that the building has a uniform settlement rate distribution and low overall strength, classifying it as a structurally stable building.

[0173] 2. If G k >0.5 and This may indicate localized uneven settlement, classifying the building as a localized anomalous type.

[0174] 3. If G k ≤0.5 and The building exhibits overall settlement or uplift, and is classified as a building with overall settlement.

[0175] 4. If G k >0.5 and The building has an uneven settlement rate distribution and a large overall settlement intensity, and is classified as a building with severe uneven settlement.

[0176] The preliminary classification results evaluate the overall stability of the building and provide an important basis for further detailed analysis.

[0177] In the second stage, this invention further analyzes the local settlement behavior of buildings initially classified as either locally anomalous or severely uneven, combining differential settlement rates and the horizontal distance between point pairs. Let building B be an example. k The PS point pair is (P i ,P j The differential settling rate is Δv ij =|v i -v j |, horizontal distance is Calculate angular distortion The classification criteria are as follows:

[0178] 1. Overall tilt judgment: When L ij When the angle distortion is greater than 10m, θ ij This represents the degree of overall tilt of the building. The building's tilt rate (i.e., maximum tilt rate) Expressed as the tilt angle per unit time (rad / yr). According to standards, the building tilt rate θ... tilt >1.0×10 -3 A rad / yr reading can indicate an abnormal tilt.

[0179] 2. Local distortion judgment: When L ij When ≤10m, the angular distortion θ ij This reflects the magnitude of localized distortion in the building. Local distortion rate (i.e., maximum distortion rate) This is expressed as the local torsion angle per unit time (rad / yr). According to the standard, the local torsion rate θ... warp >2.5×10 -3 The rad / yr ratio indicates an abnormal distortion.

[0180] In the third stage, this invention, combining the significance of overall tilt and local twisting, refines the building settlement patterns into the following three types:

[0181] 1. Both tilting and twisting are significant: If the building simultaneously satisfies the overall tilt rate (i.e., the maximum tilt rate) θ tilt >1.0×10 -3 rad / yr and local twist rate (i.e., maximum twist rate) θ warp >2.5×10 -3 rad / yr indicates that the building may be affected by complex foundation conditions or load distribution, and has high potential risks.

[0182] 2. Inclination is dominant: If the overall inclination rate (i.e., the maximum inclination rate) θ tilt >1.0×10-3 rad / yr, while the local twist rate (i.e., the maximum twist rate) θ warp ≤2.5×10 -3 rad / yr indicates that the building is mainly affected by the non-uniform distribution of foundation bearing capacity or underground activities.

[0183] 3. Twist as the dominant factor: If the local twist rate (i.e., the maximum twist rate) θ warp >2.5×10 -3 rad / yr, while the overall tilt rate (i.e., the maximum tilt rate) θ tilt ≤1.0×10 -3 rad / yr indicates that there may be foundation instability or abnormal structural stress in local areas of the building.

[0184] In the above embodiments, the stability analysis of the target building settlement point and the target net settlement rate is performed, and the analysis results are used as the assessment results of building stability. The building settlement behavior can be identified from the overall characteristics, and the building settlement pattern can be refined by combining the performance of angular distortion at different spatial scales. This provides a scientific basis for building stability assessment and safety assessment, and provides technical support for early intervention and repair of abnormal buildings.

[0185] Optionally, as another embodiment of the present invention, this invention aims to provide a systematic solution for urban building stability assessment and structural risk assessment. This system, through multi-module collaborative work, starts with the processing and optimization of time-series InSAR data, and after multi-level feature quantification analysis, ultimately achieves a refined classification of building settlement patterns. This invention utilizes time-series InSAR technology to extract large-scale surface deformation information, improves the accuracy and coverage of deformation rates through multi-track data fusion, separates pseudo-deformation factors by combining environmental data, and extracts a high-quality set of PS points related to buildings to construct settlement rate data for building units. In the feature quantification stage, this invention uses the settlement rate distribution index to quantify the overall settlement uniformity of the building, and characterizes the uneven settlement characteristics of local areas of the building through indicators such as angular distortion, comprehensively assessing the overall and local settlement status of the building. In the classification stage, based on the quantified features, the system constructs building settlement pattern classification rules from both overall and local perspectives, combining core indicators such as the Gini index, average settlement rate, differential settlement rate, and point-to-point distance to clarify the settlement behavior characteristics and potential structural risks of the building. Ultimately, building settlement patterns are categorized into three types: predominantly tilted, predominantly torsional, and both significant tilting and torsional patterns, providing a multi-dimensional interpretation method for building stability and risk assessment. This invention systematically achieves comprehensive analysis from large-scale InSAR data to individual building settlement patterns. The classification results can reflect the overall stability and local risks of buildings, providing a scientific basis and technical support for urban building health management, foundation stability assessment, and structural safety intervention. It has broad practical application value and promising prospects for promotion.

[0186] Optionally, as another embodiment of the present invention, the present invention includes a time-series InSAR data processing module, an InSAR deformation result processing module, a settlement feature quantification module, and a settlement pattern classification module, characterized in that:

[0187] The time-series InSAR data processing module is used to process the input multi-time-series SAR images. Through image registration, coherence point extraction, residual phase removal, and deformation inversion, it generates a large-scale, high-precision time series of surface deformation rate and deformation.

[0188] The InSAR deformation result processing module improves the spatiotemporal resolution through multi-track data fusion, removes pseudo-deformation caused by temperature changes through thermal deformation removal, and extracts a set of PS points related to the building by combining building outline information, and decomposes the deformation rate into vertical settlement rate.

[0189] The settlement feature quantification module is based on the extracted set of building PS points. By constructing a point-to-point dataset, it calculates the settlement rate distribution index, differential settlement rate, and angular distortion of the building, and comprehensively quantifies the overall and local settlement characteristics of the building.

[0190] The settlement pattern classification module classifies the settlement patterns of buildings systematically from the overall to the local level, based on the overall characteristics and local differences of the building, combined with features such as Gini index, average settlement rate, differential settlement rate and angular distortion.

[0191] The multi-track data fusion employs a baseline rate correction, track geometric perspective correction, and weighted strategy to integrate redundant information from different track InSAR data into a unified deformation rate field. This effectively eliminates errors caused by differences in imaging geometry between tracks and significantly improves spatiotemporal coverage and deformation measurement accuracy. The thermal deformation removal utilizes a partial least squares regression (PLSR) model to separate deformation components related to the temperature time series from the building deformation time series, generating a net settlement rate. This eliminates the interference of thermal expansion and contraction effects on the deformation rate, ensuring that building settlement characteristics truly reflect structural settlement characteristics. The building PS point extraction uses geometric spatial analysis to screen PS points overlapping with the building outline and removes low-coherence and isolated points to improve data quality. For the building unit PS point set, external PS points unrelated to the target building are removed by limiting the building outline range, ensuring the accuracy and reliability of the extraction results. The settlement feature quantification module uses the following methods:

[0192] The Gini coefficient is calculated using the set of PS points of the building to quantify the uniformity of the overall settlement rate distribution of the building. The larger the value, the more uneven the settlement distribution.

[0193] By constructing a point-pair dataset within the building, the settlement rate difference and horizontal distance between point pairs are calculated. Point-pair angular distortion is used to characterize the non-uniformity of local settlement, and the maximum angular distortion value is extracted as an important indicator of local settlement characteristics. The settlement pattern classification module performs a preliminary classification of the overall settlement characteristics of the building by combining the Gini index and the average settlement rate, specifically including:

[0194] 1. Buildings with a low Gini index and a low average settlement rate are classified as structurally stable.

[0195] 2. Buildings with a large Gini index but a small average settlement rate are classified as locally anomalous.

[0196] 3. Buildings with a low Gini index but a high average settlement rate are classified as overall settlement type;

[0197] 4. Buildings with a large Gini index and a large average settlement rate are classified as severely heterogeneous.

[0198] The settlement pattern classification module further distinguishes the local features of a building by combining differential settlement rate and point-to-point distance. By setting distance thresholds and rate difference thresholds, it determines the overall tilt and local torsion characteristics of the building, and utilizes the tilt rate threshold (θ) from building codes. tilt >1.0×10 -3 rad / yr) and the twist rate threshold (θ) warp >2.5×10 - 3 The method (rad / yr) quantifies the degree of anomalousness in building settlement behavior. The final classification results include an overall stability assessment of the building, characteristics of local uneven settlement, and a classification of settlement patterns. These results accurately reflect the overall and local settlement status of the building, providing a scientific basis for urban building stability assessment, foundation stability assessment, and structural safety risk identification. This invention can be widely applied to building settlement monitoring in complex urban environments. Through a multi-level analysis method from the overall to the local perspective, it provides comprehensive technical support and a theoretical foundation for the early identification and scientific repair of abnormal buildings.

[0199] Alternatively, as another embodiment of the present invention, the present invention adopts a modular design concept, combining large-area deformation monitoring and building unit feature analysis, and constructs a complete building settlement assessment system from data acquisition, feature extraction to pattern classification.

[0200] Optionally, as another embodiment of the present invention, the present invention realizes multi-level interpretation of building settlement patterns from the overall to the local. The classification results can not only reflect the overall settlement stability of the building, but also reveal the distribution of potential risks inside the building, providing important technical support and theoretical basis for urban building stability assessment and structural safety assessment.

[0201] Optionally, as another embodiment of the present invention, after calculating all point pairs inside the building, the present invention extracts the maximum angular distortion θ in the building. max and its corresponding point pair information. Combined with θ max The horizontal distance L between the points ij This allows for further differentiation of uneven settlement patterns. If the horizontal distance between the points of maximum angular distortion is large (e.g., L...),... ij If the angle distortion is greater than 10m, it indicates that the building is tilted overall, usually caused by uneven foundation settlement or underground activity; if the horizontal distance between the points of maximum angular distortion is small (e.g., L), it indicates that the building is tilted overall, usually caused by uneven foundation settlement or underground activity. ijIf the value is less than 10m, it indicates that the building has localized settlement distortion, which may lead to localized stress concentration and structural damage. By calculating the building settlement rate distribution index and uneven settlement characteristics, this invention comprehensively quantifies the building's settlement characteristics, providing crucial support for identifying abnormal settlement risks. These indicators not only accurately describe the overall settlement behavior of the building but also provide a scientific basis for assessing localized structural damage and potential risks.

[0202] Figure 2 This is a block diagram of a building stability assessment device provided in an embodiment of the present invention.

[0203] Alternatively, as another embodiment of the present invention, such as Figure 2 As shown, a building stability assessment device includes:

[0204] The import module is used to import multiple raw SAR images;

[0205] The differential interferometry module is used to perform differential interferometry on all the original SAR images to obtain multiple differential interferograms;

[0206] The screening and analysis module is used to screen and analyze all the differential interferograms to obtain multiple high coherence points and the initial phase time series corresponding to each of the high coherence points;

[0207] The import module is also used to import multiple terrain information corresponding to each of the high coherence points and multiple track information corresponding to each of the high coherence points;

[0208] The deformation phase analysis module is used to analyze the initial phase time series corresponding to each of the high coherence points, multiple terrain information corresponding to each of the high coherence points, and multiple orbital information corresponding to each of the high coherence points, respectively, to obtain the deformation time series corresponding to each of the high coherence points and the initial deformation rate corresponding to each of the high coherence points.

[0209] The import module is also used to import multiple temperature data corresponding to each of the high coherence points;

[0210] The rate analysis module is used to perform rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and multiple temperature data corresponding to each of the high coherence points, to obtain the thermal deformation rate corresponding to each of the high coherence points.

[0211] The difference calculation module is used to calculate the difference between the initial deformation rate corresponding to each of the high coherence points and the thermal deformation rate corresponding to each of the high coherence points, so as to obtain the initial net deformation rate corresponding to each of the high coherence points.

[0212] The import module is also used to import the projection ratio values ​​corresponding to each of the highly coherent points;

[0213] The vertical decomposition module is used to perform vertical decomposition on the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points, so as to obtain the vertical net settlement rate corresponding to each of the high coherence points.

[0214] The import module is also used to import building outline vector data;

[0215] The filtering module is used to filter out multiple target building settlement points and the target net settlement rate corresponding to each target building settlement point from all the high coherence points based on the building outline vector data and all the vertical net settlement rates.

[0216] The evaluation result acquisition module is used to perform stability analysis on all the target building settlement points and all the target net settlement rates, and use the analysis results as the evaluation results of building stability.

[0217] Optionally, another embodiment of the present invention provides a building stability assessment system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the building stability assessment method described above. This system can be a computer or similar system.

[0218] Optionally, another embodiment of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the building stability assessment method as described above.

[0219] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for assessing building stability, characterized in that, Includes the following steps: Import multiple raw SAR images, perform differential interferometry on all the raw SAR images, and obtain multiple differential interferograms; By screening and analyzing all the differential interferograms, multiple high coherence points and the initial phase time series corresponding to each high coherence point are obtained; Import multiple terrain information and multiple orbit information corresponding to each of the high coherence points, and analyze the deformation phase of the initial phase time series corresponding to each of the high coherence points, the multiple terrain information corresponding to each of the high coherence points, and the multiple orbit information corresponding to each of the high coherence points to obtain the deformation time series and the initial deformation rate corresponding to each of the high coherence points. Import multiple temperature data corresponding to each of the high coherence points, and perform rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and the multiple temperature data corresponding to each of the high coherence points to obtain the thermal deformation rate corresponding to each of the high coherence points. The difference between the initial deformation rate corresponding to each of the high coherence points and the thermal deformation rate corresponding to each of the high coherence points is calculated to obtain the initial net deformation rate corresponding to each of the high coherence points. Import the projection ratio corresponding to each of the high coherence points, and perform vertical decomposition on the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points to obtain the vertical net settlement rate corresponding to each of the high coherence points. Import building outline vector data, and based on the building outline vector data and all the vertical net settlement rates, filter out multiple target building settlement points and the target net settlement rate corresponding to each of the target building settlement points from all the high coherence points. Stability analysis of the building is performed on all the target building settlement points and all the target net settlement rates, and the analysis results are used as the evaluation results of building stability. The differential interferogram includes multiple initial pixels, initial amplitude values ​​corresponding to each initial pixel, and initial phase values ​​corresponding to each initial pixel. The process of filtering and analyzing all the differential interferograms to obtain multiple high-coherence points and the initial phase time series corresponding to each high-coherence point includes: The standard deviations of multiple initial amplitude values ​​with the same initial pixel correlation in all the differential interferograms are calculated to obtain the amplitude standard deviations with the same initial pixel correlation in all the differential interferograms. The average value of multiple initial amplitude values ​​with the same initial pixel correlation in all the differential interferograms is calculated to obtain the average amplitude value with the same initial pixel correlation in all the differential interferograms. The amplitude dispersion index of all differential interferograms with the same initial pixel correlation is obtained by calculating the amplitude standard deviation and amplitude mean of all differential interferograms with the same initial pixel correlation using the first formula. The first formula is: ， in, All differential interferograms have the same initial pixel points The amplitude dispersion index of correlation All differential interferograms have the same initial pixel points The standard deviation of the magnitude of the correlation All differential interferograms have the same initial pixel points The mean magnitude of the correlation; If the amplitude dispersion index is less than the preset amplitude threshold, then all the initial pixels corresponding to the amplitude dispersion index are taken as candidate pixels, and the initial phase value corresponding to the candidate pixels is taken as a candidate phase value, thereby obtaining multiple candidate pixels and multiple candidate phase values ​​with the same candidate pixel correlation in all the differential interferograms. The temporal coherence coefficients of the same candidate pixel correlation in all the differential interferograms are obtained by calculating the candidate phase values ​​with the same candidate pixel correlation in all the differential interferograms using the second equation. The second equation is: ， in, For all differential interferograms with the same candidate pixels The temporal coherence coefficient of correlation, The total number of differential interferograms, For the first The difference interferograms contain the same candidate pixels. Candidate phase values ​​for correlation; If the temporal coherence coefficient is greater than the preset coherence coefficient threshold, then all the candidate pixels corresponding to the temporal coherence coefficient are taken as pixels to be processed, and the candidate phase values ​​corresponding to the pixels to be processed are taken as phase values ​​to be processed, thereby obtaining multiple pixels to be processed and multiple phase values ​​to be processed that have the same correlation with the pixels to be processed in all the differential interferograms. The probability of each pixel to be processed is calculated by using a pre-constructed probability distribution model to obtain the pixel probability value corresponding to each pixel to be processed. If the pixel probability value is greater than a preset probability threshold, then the pixel to be processed corresponding to the pixel probability value is taken as the processed pixel, and the phase value to be processed corresponding to the processed pixel is taken as the target phase value, thereby obtaining multiple processed pixels and the target phase value corresponding to each processed pixel. All processed pixels in the differential interferograms are grouped together as high coherence points, thereby obtaining multiple high coherence points and multiple target phase values ​​corresponding to each high coherence point; An initial phase time series corresponding to each of the high coherence points is constructed by using multiple target phase values ​​corresponding to each of the high coherence points.

2. The building stability assessment method according to claim 1, characterized in that, The process of analyzing the deformation phase of the initial phase time series corresponding to each of the high coherence points, multiple terrain information corresponding to each of the high coherence points, and multiple orbital information corresponding to each of the high coherence points to obtain the deformation time series and the initial deformation rate corresponding to each of the high coherence points includes: The network flow algorithm is used to perform phase unwrapping processing on the initial phase time series corresponding to each of the high coherence points to obtain the unwrapped phase time series corresponding to each of the high coherence points. The unwrapped phase time series corresponding to each of the high coherence points are filtered to obtain the deformation phase series corresponding to each of the high coherence points. A linear deformation model is constructed using all the aforementioned deformation phase sequences; The linear deformation model is fitted using the least squares algorithm to obtain the deformation time series corresponding to each of the high coherence points and the initial deformation rate corresponding to each of the high coherence points.

3. The building stability assessment method according to claim 1, characterized in that, The process of performing rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and multiple temperature data corresponding to each of the high coherence points to obtain the thermal deformation rate corresponding to each of the high coherence points includes: By fitting the deformation time series corresponding to each of the high coherence points and multiple temperature data corresponding to each of the high coherence points using a pre-constructed partial least squares regression model, the thermal deformation components corresponding to each of the high coherence points are obtained. The thermal deformation rate corresponding to each of the highly coherent points is obtained by taking the derivative of the thermal deformation component.

4. The building stability assessment method according to claim 1, characterized in that, The process of vertically decomposing the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points to obtain the vertical net settlement rate corresponding to each of the high coherence points includes: By vertically decomposing the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points using the third equation, the vertical net settlement rate corresponding to each of the high coherence points is obtained. The third equation is: ， in, For the first The vertical net settlement rate corresponding to each high coherence point For the first The initial net deformation rate corresponding to each high coherence point For the first The projection ratio corresponding to each highly coherent point.

5. The building stability assessment method according to claim 1, characterized in that, The process of selecting multiple target building settlement points and the target net settlement rate corresponding to each target building settlement point from all the high coherence points based on the building outline vector data and all the vertical net settlement rates includes: S71: Based on the judgment criteria and the building outline vector data, multiple initial building deformation points are selected from all the highly coherent points. The judgment criteria are: ， in, For the first An initial building deformation point, For the first The X-axis coordinates of the high coherence points for The Y-axis coordinates of the high coherence points This is the vector data of the building outline; S72: Calculate the distance between each of the initial building deformation points and any remaining initial building deformation point to obtain the distances between multiple initial neighboring points corresponding to each of the initial building deformation points; S73: Filter the minimum value of the distances between multiple initial neighboring points corresponding to each of the initial building deformation points to obtain the target neighboring point distances corresponding to each of the initial building deformation points; S74: When the target neighbor distance is less than or equal to the preset neighbor distance threshold, the initial building deformation point corresponding to the target neighbor distance is used as the processed building deformation point, thereby obtaining multiple processed building deformation points. S75: Count the number of all the processed building deformation points to obtain the total number of processed building deformation points; S76: Determine whether the total number of building deformation points after processing is greater than or equal to the preset number of building deformation points. If not, re-import the building outline vector data and return to S71. If yes, use the processed building deformation points as target building settlement points to obtain multiple target building settlement points, and use the vertical net settlement rate corresponding to the target building settlement points as the target net settlement rate to obtain the target net settlement rate corresponding to each target building settlement point.

6. The building stability assessment method according to any one of claims 1 to 5, characterized in that, The process of performing stability analysis on all target building settlement points and all target net settlement rates, and using the analysis results as the assessment results of building stability, includes: S81: The building deformation rate distribution index is obtained by calculating the building deformation rate distribution index for all target building settlement points and all target net settlement rates using the fourth formula. The fourth formula is: ， in, This is the building deformation rate distribution index. The total number of settlement points of the target building. For the first The target net settlement rate corresponding to each target building settlement point For the first The target net settlement rate corresponding to each target building settlement point; S82: Calculate the average net settlement rate of all the targets to obtain the average deformation rate; S83: Determine whether the building deformation rate distribution index is less than or equal to the preset building deformation rate distribution threshold. If yes, proceed to S84; otherwise, proceed to S85. S84: Determine whether the average deformation rate is less than or equal to a preset deformation rate threshold. If yes, use the first preset evaluation result as the evaluation result of building stability; otherwise, use the second preset evaluation result as the evaluation result of building stability. S85: The fifth formula is used to calculate the horizontal distances to each of the target building settlement points and any remaining target building settlement point, respectively, to obtain multiple horizontal distances corresponding to each of the target building settlement points. The fifth formula is: ， in, For the first The target building settlement point and the first Horizontal distance between the settlement points of the target buildings For the first The X-axis coordinates of the settlement points of the target building. For the first The X-axis coordinates of the settlement points of the target building. For the first The Y-axis coordinates of the settlement points of the target buildings. For the first Y-axis coordinates of the settlement points of the target buildings; S86: The differential deformation rate is calculated by using the sixth equation to calculate the differential deformation rate corresponding to each of the stated target net settlement rates and any remaining target net settlement rate, respectively. The sixth equation is: ， in, For the first The target building settlement point and the first Differential deformation rates at settlement points of target buildings For the first The target net settlement rate corresponding to each target building settlement point For the first The target net settlement rate corresponding to each target building settlement point; S87: Determine whether each horizontal distance is greater than a preset horizontal distance threshold. If yes, proceed to S88; otherwise, proceed to S89. S88: The tilt rate corresponding to each horizontal distance is calculated by using the seventh formula to calculate the tilt rate for each horizontal distance and the differential deformation rate corresponding to each horizontal distance. The seventh formula is: ， in, , in, For the first The target building settlement point and the first The tilt rate of the target building settlement point For the first The target building settlement point and the first The angular distortion value of the settlement point of the target building. To preset the first sampling time, For the first The target building settlement point and the first Differential deformation rates at settlement points of target buildings For the first The target building settlement point and the first The horizontal distance between the settlement points of the target buildings; S89: The torsion rate corresponding to each horizontal distance is calculated by using the eighth formula to calculate the torsion rate for each horizontal distance and the differential deformation rate corresponding to each horizontal distance. The eighth formula is: ， in, , in, For the first The target building settlement point and the first The torsion rate of the settlement point of the target building For the first The target building settlement point and the first The angular distortion value of the settlement point of the target building. To preset the second sampling time, For the first The target building settlement point and the first Differential deformation rates at settlement points of target buildings For the first The target building settlement point and the first The horizontal distance between the settlement points of the target buildings; S810: Filter out the maximum value among all the tilt rates to obtain the maximum tilt rate; S811: Filter out the maximum value among all the stated twist rates to obtain the maximum twist rate; S812: When the average deformation rate is less than or equal to the preset deformation rate threshold, and the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, then the third preset evaluation result is taken as the evaluation result of building stability. S813: When the average deformation rate is less than or equal to the preset deformation rate threshold, and the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is less than or equal to the preset torsion rate threshold, then the fourth preset evaluation result is taken as the evaluation result of building stability. S814: When the average deformation rate is less than or equal to the preset deformation rate threshold, the maximum tilt rate is less than or equal to the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, then the fifth preset evaluation result is taken as the evaluation result of building stability. S815: When the average deformation rate is greater than the preset deformation rate threshold, the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, the sixth preset evaluation result shall be used as the evaluation result of building stability. S816: When the average deformation rate is greater than the preset deformation rate threshold, the maximum tilt rate is greater than the preset tilt rate threshold, and the maximum torsion rate is less than or equal to the preset torsion rate threshold, the seventh preset evaluation result shall be used as the evaluation result of building stability. S817: When the average deformation rate is greater than the preset deformation rate threshold, and the maximum tilt rate is less than or equal to the preset tilt rate threshold, and the maximum torsion rate is greater than the preset torsion rate threshold, then the eighth preset evaluation result is taken as the evaluation result of building stability.

7. A building stability assessment device, characterized in that, include: The import module is used to import multiple raw SAR images; The differential interferometry module is used to perform differential interferometry on all the original SAR images to obtain multiple differential interferograms; The screening and analysis module is used to screen and analyze all the differential interferograms to obtain multiple high coherence points and the initial phase time series corresponding to each of the high coherence points; The import module is also used to import multiple terrain information corresponding to each of the high coherence points and multiple track information corresponding to each of the high coherence points; The deformation phase analysis module is used to analyze the initial phase time series corresponding to each of the high coherence points, multiple terrain information corresponding to each of the high coherence points, and multiple orbital information corresponding to each of the high coherence points, respectively, to obtain the deformation time series corresponding to each of the high coherence points and the initial deformation rate corresponding to each of the high coherence points. The import module is also used to import multiple temperature data corresponding to each of the high coherence points; The rate analysis module is used to perform rate analysis on the deformation time series corresponding to each of the high coherence points, the initial deformation rate corresponding to each of the high coherence points, and multiple temperature data corresponding to each of the high coherence points, to obtain the thermal deformation rate corresponding to each of the high coherence points. The difference calculation module is used to calculate the difference between the initial deformation rate corresponding to each of the high coherence points and the thermal deformation rate corresponding to each of the high coherence points, so as to obtain the initial net deformation rate corresponding to each of the high coherence points. The import module is also used to import the projection ratio values ​​corresponding to each of the highly coherent points; The vertical decomposition module is used to perform vertical decomposition on the initial net deformation rate corresponding to each of the high coherence points and the projection ratio corresponding to each of the high coherence points, so as to obtain the vertical net settlement rate corresponding to each of the high coherence points. The import module is also used to import building outline vector data; The filtering module is used to filter out multiple target building settlement points and the target net settlement rate corresponding to each target building settlement point from all the high coherence points based on the building outline vector data and all the vertical net settlement rates. The evaluation result acquisition module is used to perform stability analysis on all the target building settlement points and all the target net settlement rates, and use the analysis results as the evaluation results of building stability. The differential interferogram includes multiple initial pixels, initial amplitude values ​​corresponding to each initial pixel, and initial phase values ​​corresponding to each initial pixel. The screening and analysis module is specifically used for: The standard deviations of multiple initial amplitude values ​​with the same initial pixel correlation in all the differential interferograms are calculated to obtain the amplitude standard deviations with the same initial pixel correlation in all the differential interferograms. The average value of multiple initial amplitude values ​​with the same initial pixel correlation in all the differential interferograms is calculated to obtain the average amplitude value with the same initial pixel correlation in all the differential interferograms. The amplitude dispersion index of all differential interferograms with the same initial pixel correlation is obtained by calculating the amplitude standard deviation and amplitude mean of all differential interferograms with the same initial pixel correlation using the first formula. The first formula is: ， in, All differential interferograms have the same initial pixel points The amplitude dispersion index of correlation All differential interferograms have the same initial pixel points The standard deviation of the magnitude of the correlation All differential interferograms have the same initial pixel points The mean magnitude of the correlation; If the amplitude dispersion index is less than the preset amplitude threshold, then all the initial pixels corresponding to the amplitude dispersion index are taken as candidate pixels, and the initial phase value corresponding to the candidate pixels is taken as a candidate phase value, thereby obtaining multiple candidate pixels and multiple candidate phase values ​​with the same candidate pixel correlation in all the differential interferograms. The temporal coherence coefficients of the same candidate pixel correlation in all the differential interferograms are obtained by calculating the candidate phase values ​​with the same candidate pixel correlation in all the differential interferograms using the second equation. The second equation is: ， in, For all differential interferograms with the same candidate pixels The temporal coherence coefficient of correlation, The total number of differential interferograms, For the first The difference interferograms contain the same candidate pixels. Candidate phase values ​​for correlation; If the temporal coherence coefficient is greater than the preset coherence coefficient threshold, then all the candidate pixels corresponding to the temporal coherence coefficient are taken as pixels to be processed, and the candidate phase values ​​corresponding to the pixels to be processed are taken as phase values ​​to be processed, thereby obtaining multiple pixels to be processed and multiple phase values ​​to be processed that have the same correlation with the pixels to be processed in all the differential interferograms. The probability of each pixel to be processed is calculated by using a pre-constructed probability distribution model to obtain the pixel probability value corresponding to each pixel to be processed. If the pixel probability value is greater than a preset probability threshold, then the pixel to be processed corresponding to the pixel probability value is taken as the processed pixel, and the phase value to be processed corresponding to the processed pixel is taken as the target phase value, thereby obtaining multiple processed pixels and the target phase value corresponding to each processed pixel. All processed pixels in the differential interferograms are grouped together as high coherence points, thereby obtaining multiple high coherence points and multiple target phase values ​​corresponding to each high coherence point; An initial phase time series corresponding to each of the high coherence points is constructed by using multiple target phase values ​​corresponding to each of the high coherence points.

8. A building stability assessment system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the building stability assessment method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the building stability assessment method as described in any one of claims 1 to 6.

Images