Highway slope monitoring method and device based on multi-platform insar cooperation

Abstract

This invention provides a method and apparatus for highway slope monitoring based on multi-platform InSAR collaboration, belonging to the field of highway slope monitoring technology. The method includes: extracting surface deformation information under the rising and falling orbits from the time-series SAR satellite images of the target area corresponding to the highway to be monitored, performing phase calculation to obtain two-dimensional deformation information along the radar line of sight; calculating the two-dimensional deformation information within a determined visibility differentiation map to obtain the deformation characteristics of the target area in the vertical and slope directions, respectively, to identify road sections with potential slope hazards; obtaining an interferometric phase map based on UAV SAR images acquired from the road sections with potential slope hazards, inverting the deformation characteristics in the road sections with potential slope hazards, and using the inverted deformation time-series information to predict slope risk, thereby obtaining the slope monitoring results of the highway to be monitored. This invention addresses the limitations of single-platform InSAR in highway monitoring and the insufficient dimensionality of monitored highway deformation information.

Description

Technical Field

[0001] This invention relates to the field of highway slope monitoring technology, specifically to a highway slope monitoring method and device based on multi-platform InSAR collaboration. Background Technology

[0002] Highway safety has long been threatened by geological disasters such as slope landslides. Traditional slope monitoring methods mainly rely on manual inspections and distributed sensor deployment, which have problems such as low monitoring efficiency, high personnel safety risks, limited coverage, and difficulty in achieving early warning and continuous dynamic monitoring.

[0003] In recent years, Interferometric Synthetic Aperture Radar (InSAR) technology has been widely used in deformation monitoring due to its advantages such as millimeter-level accuracy, wide coverage, and immunity to weather conditions. However, single-platform InSAR technology has limitations. For example, while spaceborne InSAR has a wide coverage area, its spatial resolution and timeliness in monitoring small deformations are limited. Emerging UAV-borne InSAR, while possessing high mobility and flexible observation capabilities, still faces technical challenges in terms of the stability of Synthetic Aperture Radar (SAR) data acquisition and processing complexity. Furthermore, InSAR technology suffers from insufficient deformation information dimensions in highway monitoring. Its highway measurement data only provides one-dimensional deformation information along the radar line of sight, making it difficult to comprehensively and accurately reflect the motion characteristics of highway slopes in different directions, thus limiting its refined application in monitoring complex highway terrain slopes. Summary of the Invention

[0004] In view of this, it is necessary to provide a method and device for highway slope monitoring based on multi-platform InSAR collaboration, so as to solve the problems of the limitations of single-platform InSAR in highway monitoring and the insufficient dimensions of highway deformation information monitored.

[0005] To address the aforementioned problems, this invention provides a highway slope monitoring method based on multi-platform InSAR collaboration, comprising: The surface deformation information under the rising and falling orbit is extracted from the rising and falling orbit time-series SAR satellite image of the target area corresponding to the highway to be tested, and the surface deformation phase is calculated to obtain the two-dimensional deformation information along the radar line of sight under the rising and falling orbit time-series SAR satellite image. The visible area is identified by the time-series SAR satellite imagery of the ascending and descending orbits to obtain a visibility distinction map of the highway to be measured. Within the visibility distinction map, the deformation information of the two-dimensional deformation is calculated to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively. Based on the deformation characteristics, identify the sections of the highway with potential slope hazards. Based on UAV SAR images obtained from the road section with potential slope hazards, an interferometric phase map with deformation phase is obtained, and the deformation characteristics in the road section with potential slope hazards are inverted according to the interferometric phase map to obtain deformation time series information. The deformation time series information is used to predict slope risk, and the slope monitoring results of the highway to be tested are obtained.

[0006] In one possible implementation, the surface deformation information includes permanent scatterer points and distributed scatterer points in the target area, and the surface deformation phase calculation process for the permanent scatterer points and the distributed scatterer points includes: The permanent scatterer points and the distributed scatterer points are used as nodes to construct a Delaunay triangulation, and differential interference phase calculation is performed on the edges in the Delaunay triangulation to obtain the absolute phase of the nodes. Obtain digital elevation model data and GACOS atmospheric correction data for the target area; Based on the terrain interferometric phase simulated by the digital elevation model data, the absolute phase is unwrapped, and atmospheric delay phase removal is performed on the unwrapped continuous phase field based on atmospheric correction data to obtain the deformation phase of the target area. Spatiotemporal parameters are inverted on the deformation phase to obtain two-dimensional deformation information along the radar line of sight in the ascending and descending orbit time-series SAR satellite image.

[0007] In one possible implementation, the ascending and descending orbit time-series SAR satellite imagery includes both ascending orbit time-series SAR satellite imagery and descending orbit time-series SAR satellite imagery of the target area. The step of performing visible area identification on the ascending and descending orbit time-series SAR satellite imagery to obtain a visibility differentiation map of the highway to be measured includes: Based on the incident angle, terrain slope, and slope direction of the ascending and descending orbit time-series SAR satellite images, the R-index is calculated for ascending and descending orbits respectively. The visible and invisible areas of the target region are determined based on the R index; The target area that is divided into visible areas in the time-series SAR satellite imagery of the ascending and descending orbits is identified as the dual-track visible area of ​​the highway to be measured. The target area that is divided into a visible area in the ascending-descending orbit time-series SAR satellite image or the descending orbit time-series SAR satellite image is determined as the single-track visible area of ​​the highway to be measured; The target areas that are all classified as out-of-view areas in the time-series SAR satellite images of the ascending and descending orbits are identified as the out-of-view areas of the highway to be measured. A visibility differentiation map of the highway under test is constructed based on the dual-track visible area, the single-track visible area, and the invisible area.

[0008] In one possible implementation, the step of performing deformation calculation on the two-dimensional deformation information within the visibility differentiation map to obtain the deformation features of the target area in the vertical direction and the slope direction, respectively, includes: The deformation of the slope along the radar line of sight in the two-dimensional deformation information is decomposed into multiple deformation components, including vertical deformation components, slope surface deformation components, and ridge orientation deformation components. In the dual-track visible area of ​​the visibility differentiation map, the deformation of the lifting and lowering tracks in the two-dimensional deformation information is calculated based on the deformation components under the lifting and lowering tracks to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively. In the single-track visible area of ​​the visibility differentiation map, the deformation of the rising or falling track in the two-dimensional deformation information is calculated based on the deformation component under the rising or falling track to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively.

[0009] In one possible implementation, determining the slope hazard section of the highway to be tested based on the deformation characteristics includes: Based on the deformation features, a deformation feature vector is constructed, and based on the deformation feature vector, the kernel density corresponding to each deformation point in the target region is calculated, and a corresponding deformation heat map is generated. Obtain the route vector map and slope deformation rate raster map of the highway to be tested, and determine the preset buffer area on both sides of the highway route in the route vector map. Extract the target heat map from the deformation heat map in the preset buffer area. The region with a kernel density greater than the kernel density threshold in the target hotspot map is identified as a deformation clustering region, and the deformation clustering region is spatially superimposed with the slope deformation rate raster map to obtain the target clustering region. When the deformation hotspot level of the target cluster area exceeds the level threshold and the slope deformation rate exceeds the rate threshold, the road route in the target cluster area is identified as a section of the road with potential slope hazards on the road to be tested.

[0010] In one possible implementation, obtaining an interferometric phase map with deformation phase based on UAV SAR imagery acquired from the road section with potential slope hazards includes: An initial interferometric phase map was obtained based on UAV SAR images acquired from the road section with potential slope hazards. Based on the sensor spatial position coordinates of the UAV SAR image at the imaging time, the flat-ground phase component is calculated, and the flat-ground phase component is removed from the initial interferometric phase map to obtain the target coherent phase map; High coherence points with temporal coherence coefficients greater than a coefficient threshold are selected from the pixels of the target coherent phase map; Using the planar coordinates of the highly coherent points as independent variables and the phase value of the highly coherent points in the target coherent phase diagram as dependent variables, a quadratic polynomial model is constructed. The quadratic polynomial model characterizes the orbital trend phase caused by residual spatial baseline errors. The polynomial coefficients of the quadratic polynomial model are analyzed, and the optimal estimate of the trajectory trend phase is calculated based on the polynomial coefficients. Based on the optimal estimate, an orbital trend correction surface of the same size as the initial interferometric phase map is generated, and the orbital trend correction surface is removed from the target coherent phase map to obtain an interferometric phase map with deformed phase.

[0011] In one possible implementation, the step of performing slope risk prediction on the deformation time series information to obtain the slope monitoring results of the highway under test includes: Based on the deformation time series information, long-term and short-term deformation characteristics of the slope hazard section are extracted; When the warning level of the slope hazard section is lower than the preset standard warning level and the duration of the warning level being lower than the preset standard warning level is greater than a time threshold, a first weight is assigned to each long-term deformation feature, and a second weight is set for each short-term deformation feature, wherein the first weight is greater than the second weight. When the warning level of the slope hazard section exceeds the preset standard warning level, a first weight is assigned to each long-term deformation feature, and a second weight is set for each short-term deformation feature, wherein the second weight is greater than the first weight. The highway slope risk score is calculated based on the short-term deformation feature, the long-term deformation feature, the first weight, and the second weight, wherein the weight assigned to each short-term deformation feature or the long-term deformation feature is negatively correlated with the feature correlation. The risk level of the highway slope is determined by the preset highway slope risk level threshold, and the determined highway slope risk level is used as the slope monitoring result of the highway to be tested.

[0012] This invention also provides a highway slope monitoring device based on multi-platform InSAR collaboration, comprising: The two-dimensional deformation calculation module is used to extract the surface deformation information under the ascending and descending orbits of the time-series SAR satellite images of the target area corresponding to the highway under test, and to perform surface deformation phase calculation to obtain the two-dimensional deformation information along the radar line of sight under the time-series SAR satellite images of the ascending and descending orbits. The deformation feature extraction module is used to identify the visible area of ​​the ascending and descending orbit time-series SAR satellite image to obtain a visibility distinction map of the highway to be tested. Within the visibility distinction map, the two-dimensional deformation information is subjected to deformation calculation to obtain the deformation features of the target area in the vertical direction and the slope direction, respectively. The slope hazard identification module is used to identify the slope hazard sections of the highway to be tested based on the deformation characteristics. The deformation feature inversion module is used to obtain an interferometric phase map with deformation phase based on UAV SAR images acquired from the road section with potential slope hazards, and to invert the deformation features in the road section with potential slope hazards based on the interferometric phase map to obtain deformation time series information. The slope risk prediction module is used to predict slope risk based on the deformation time series information and obtain the slope monitoring results of the highway to be tested.

[0013] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a program; the processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of the above-described highway slope monitoring method based on multi-platform InSAR collaboration.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described highway slope monitoring method based on multi-platform InSAR collaboration.

[0015] The beneficial effects of adopting the above implementation method are as follows: The highway slope monitoring method and device based on multi-platform InSAR collaboration provided by this invention, firstly, utilizes the large coverage of satellite InSAR to obtain deformation features in the vertical and slope directions based on the ascending and descending orbit time-series SAR satellite images of the target area of ​​the highway to be monitored, thereby initially locating the road sections with potential slope hazards in the highway to be monitored. Furthermore, UAV SAR images are used to focus on the road sections with potential slope hazards, and deformation features with temporal characteristics are stably extracted for slope risk prediction, resulting in the slope monitoring results of the highway to be monitored. This collaborative monitoring method combining satellite InSAR and UAV InSAR not only takes into account the advantages of both InSAR platforms and overcomes the limitations of single-platform InSAR in highway monitoring, but also achieves seamless connection from large-scale highway network area screening to local precise monitoring. Secondly, by analyzing the two-dimensional deformation information in the radar line-of-sight direction to extract the deformation features of the target area in the vertical and slope directions, it can comprehensively and realistically reflect the movement characteristics of the highway slope in different directions such as vertical and slope, overcoming the deficiency of insufficient one-dimensional deformation information and improving the accuracy of monitoring complex highway terrain slopes. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating the highway slope monitoring method based on multi-platform InSAR collaboration provided by this invention; Figure 2 A schematic diagram illustrating the principle of satellite InSAR positioning of road sections with potential slope hazards provided by this invention; Figure 3 A schematic diagram illustrating the principle of UAV InSAR prediction of highway slope risk provided by this invention; Figure 4 A schematic diagram of the structure of the highway slope monitoring device based on multi-platform InSAR collaboration provided by the present invention; Figure 5 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] In the description of the embodiments of this application, unless otherwise stated, "a plurality of" means two or more.

[0020] In this embodiment of the invention, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, apparatus, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product or device.

[0021] The naming or numbering of steps in the embodiments of the present invention does not mean that the steps in the method flow must be executed in the time / logical order indicated by the naming or numbering. The execution order of the named or numbered process steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effect can be achieved.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] The highway slope monitoring method and device based on multi-platform InSAR collaboration provided by this invention can be applied to slope monitoring scenarios in various highway terrains. The executing entity can be various servers, terminals, or remote cloud devices. It extracts and uploads time-series SAR satellite images of the target area corresponding to the highway under test using satellite InSAR technology. Then, it uses the multi-platform InSAR collaborative highway slope monitoring method and device provided by this invention to process the time-series SAR satellite images, first locating the potentially hazardous sections of the highway slope under test. Next, it acquires and uploads UAV SAR images of these hazardous sections using UAV InSAR technology. Finally, it uses the multi-platform InSAR collaborative highway slope monitoring method and device provided by this invention to process the UAV SAR images, obtaining the slope monitoring results for the highway under test. These monitoring results are used for highway monitoring feedback, serving as a theoretical basis for highway management and maintenance.

[0024] The following section details the highway slope monitoring method based on multi-platform InSAR collaboration provided by this invention.

[0025] Figure 1 The highway slope monitoring method based on multi-platform InSAR collaboration provided by this invention, such as Figure 1 As shown, the highway slope monitoring method based on multi-platform InSAR collaboration can be implemented through the following steps 101 to 105, which are explained in detail below.

[0026] Step 101: Extract the surface deformation information under the rising and falling orbit from the rising and falling orbit time-series SAR satellite image of the target area corresponding to the highway to be tested, and perform surface deformation phase calculation to obtain two-dimensional deformation information along the radar line of sight under the rising and falling orbit time-series SAR satellite image.

[0027] Here, during satellite InSAR data acquisition, it is necessary to cover a certain range of the road area to be measured (i.e., the target area). Specifically, this involves acquiring time-series SAR satellite images covering a 1km radius on both sides of the road's centerline. These time-series SAR satellite images include SAR satellite images acquired by the satellite at both the ascending and descending observation angles. Furthermore, SAR satellite images have a temporal sequence, with each image having a corresponding timestamp. In addition, it is also necessary to acquire Digital Elevation Model (DEM) data and atmospheric correction data corresponding to the target area. The atmospheric correction data specifically uses GACOS satellite observation data to provide spatial distribution information of global atmospheric chemical composition.

[0028] See Figure 2 In data processing, the first aspect involves using time-series SAR satellite imagery of ascending and descending orbits, digital elevation model data, and atmospheric correction data to extract deformation information from the ascending and descending orbits. This involves using synthetic aperture radar (SAR) satellites at two different observation angles (ascending and descending orbits) to determine surface deformation data, performing surface deformation phase calculations, and obtaining two-dimensional deformation information—specifically, the ascending and descending orbit deformation information of the target area. The second aspect involves using time-series SAR satellite imagery of ascending and descending orbits to perform visibility zoning of the road area under test, distinguishing between dual-track and single-track visibility areas for subsequent two-dimensional slope deformation calculations.

[0029] In one possible implementation, the surface deformation information includes permanent scatterer points and distributed scatterer points in the target area, and the surface deformation phase calculation process for the permanent scatterer points and the distributed scatterer points can be implemented through the following steps 1011 to 1013, which are described in detail below.

[0030] Step 1011: Construct a Delaunay triangulation using permanent scatterer points and distributed scatterer points as nodes, and perform differential interference phase calculation on the edges of the Delaunay triangulation to obtain the absolute phase of the nodes.

[0031] Here, for ascending and descending orbit time-series SAR satellite imagery, a method combining amplitude deviation index (ADI) and phase stability analysis is used to identify highly coherent permanent scatterer (PS) points in the target region. Simultaneously, the FaSHPS algorithm is employed for homogeneous pixel identification and coherence estimation in the ascending and descending orbit time-series SAR satellite imagery, and distributed scatterer (DS) points in the target region are selected based on the number of homogeneous pixels and their coherence. Of course, the extraction of PS and DS points is performed separately for ascending and descending orbits; the following explanation focuses on the ascending orbit scenario.

[0032] Specifically, a Delaunay triangulation is constructed using permanent and distributed scattering points as nodes. Each node has a corresponding interference phase. Therefore, for the edges of the Delaunay triangulation, the differential interference phase between two nodes can be calculated, and joint solution is achieved through weighted least squares adjustment to update the interference phase of each node and optimize the absolute phase of each node.

[0033] Step 1012: Obtain digital elevation model data and GACOS atmospheric correction data for the target area; perform phase unwrapping on the absolute phase based on the terrain interferometric phase simulated by the digital elevation model data, and perform atmospheric delay phase removal on the unwrapped continuous phase field based on the atmospheric correction data to obtain the deformation phase of the target area.

[0034] Here, we introduce digital elevation model (DEM) data and GACOS atmospheric correction data acquired synchronously during the acquisition of time-series SAR satellite images in ascending and descending orbits. The absolute phase is unwrapped using the DEM. This process first simulates the terrain interferometry phase of the target area based on the DEM data, then removes the terrain interferometry phase from the absolute phase. Next, a minimum cost flow algorithm is used for phase unwrapping to obtain the continuous phase field of the target area. Using the GACOS atmospheric correction data, the image coordinates can be resampled to the coordinate system of the time-series SAR satellite images in ascending and descending orbits via interpolation. Then, atmospheric delay phase is removed pixel-by-pixel from the unwrapped continuous phase field, thereby eliminating atmospheric interference and obtaining the deformation phase of the target area.

[0035] Step 1013: Perform spatiotemporal parameter inversion on the deformation phase to obtain two-dimensional deformation information along the radar line of sight in the ascending and descending orbit time-series SAR satellite image.

[0036] Here, a parameterized model including linear deformation, nonlinear deformation, elevation residuals and noise is first constructed. Then, the spatiotemporal parameters are inverted using least squares or singular value decomposition algorithms. The deformation phase is separated from systematic error components such as topographic residuals, atmospheric residuals, and orbital errors, and finally, a high-precision surface deformation phase is obtained.

[0037] For time-series SAR satellite imagery in descending orbit, the surface deformation phase is calculated using the same method described above, ultimately yielding a high-precision surface deformation phase. Based on this surface deformation phase, information such as the deformation rate and temporal deformation along the radar line of sight in the ascending and descending orbit time-series SAR satellite imagery can be further determined, thereby calculating the slope deformation along the radar line of sight and the ascending and descending orbit deformation pixel by pixel.

[0038] Therefore, through the above-mentioned process of surface deformation phase calculation, the two-dimensional deformation information along the radar line of sight of the rising and falling orbit time-series SAR satellite image is finally obtained, specifically including the deformation of the slope along the radar line of sight and the deformation of the rising and falling orbit.

[0039] In this embodiment of the invention, when calculating the surface deformation phase of time-series SAR satellite imagery, the absolute phase of the surface is determined by extracting permanent and distributed scatterer points, enabling millimeter-level precision in surface deformation monitoring and significantly improving the spatial coverage density of monitoring points. Furthermore, by using digital elevation model data and atmospheric correction data to correct the phase, the influence of elevation and atmosphere on surface deformation monitoring can be eliminated, ensuring the accuracy of surface deformation information extraction.

[0040] Step 102: Perform visible area identification on the ascending and descending orbit time-series SAR satellite imagery to obtain a visibility distinction map of the highway to be measured. Within the visibility distinction map, perform deformation calculation on the two-dimensional deformation information to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively.

[0041] Here, due to the difference in observation angles between ascending and descending orbits, it is necessary to perform visibility zoning on the time-series SAR satellite images. Visibility can be determined by calculating the R-index. During the visibility zone identification process, the R-index is calculated to determine the visible and invisible zones of the target area, resulting in a visibility classification map of the road under test.

[0042] In one possible implementation, visual area identification is performed on the ascending and descending orbit time-series SAR satellite imagery to obtain a visibility differentiation map of the highway under test. This can be achieved in the following ways, which are explained in detail below.

[0043] First, based on the incident angle, terrain slope, and slope direction of the time-series SAR satellite images in ascending and descending orbits, the R-index is calculated for both ascending and descending orbits, as shown in the following formula: (1) in, Indicates the angle of incidence. Indicates the slope of the terrain. Indicates the direction of the slope. This represents the R-index.

[0044] Next, the visible and invisible areas of the target region are determined based on the R index.

[0045] Here, when the R-index of a certain region satisfies In the case that, or the R exponent satisfies and < In the case of a certain condition, the area can be designated as the visible area of ​​the target area; otherwise, it can be designated as an invisible area.

[0046] Furthermore, target areas that are both visible in the ascending and descending orbit time-series SAR satellite images are identified as the dual-track visible areas of the highway to be measured; target areas that are visible in either the ascending or descending orbit time-series SAR satellite images are identified as the single-track visible areas of the highway to be measured; and target areas that are not visible in both the ascending and descending orbit time-series SAR satellite images are identified as the non-visible areas of the highway to be measured.

[0047] Here, the ascending and descending orbit time-series SAR satellite imagery includes both ascending orbit time-series SAR satellite imagery and descending orbit time-series SAR satellite imagery of the target area. Based on the visible and invisible areas defined in ascending and descending orbits respectively, the target area can be further divided into dual-track visible area, single-track visible area, and dual-track invisible area.

[0048] Dual-track visibility area represents a region that is visible in both ascending and descending orbits. That is, a target area that is visible in both ascending and descending orbit time-series SAR satellite imagery can be identified as a dual-track visibility area.

[0049] A single-track visible area is a region that is visible only in either ascending or descending orbit. In other words, a target area that is visible in both ascending and descending time-series SAR satellite imagery can be identified as a single-track visible area.

[0050] The dual-track blind zone is a region that is classified as a blind zone in both ascending and descending orbits. That is, a target region that is classified as a blind zone in both ascending and descending orbit time-series SAR satellite imagery can be identified as a dual-track blind zone.

[0051] Finally, a visibility differentiation map of the highway under test is constructed based on the dual-track visible area, the single-track visible area, and the invisible area.

[0052] Therefore, based on the above division method, the target area of ​​the highway to be tested can be divided into a dual-track visible area, a single-track visible area, and a dual-track non-visible area. These areas are then combined to form a visibility zoning map corresponding to the highway to be tested.

[0053] In this embodiment of the invention, by identifying the visible area of ​​the target area of ​​the highway under test, it is possible to effectively utilize the deformation information of satellite imagery under the rising and falling orbits to achieve deformation calculation in subsequent deformation calculation, thereby eliminating the impact of the visual blind spots of satellite imagery under different observation angles on monitoring.

[0054] Furthermore, within the visibility differentiation map, deformation calculation is performed on the two-dimensional deformation information to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively.

[0055] Since two-dimensional deformation information represents deformation along the line of sight of satellite radar, deformation information from a single track cannot truly reflect the actual deformation of the roadside slope. Therefore, it is necessary to combine the deformation data of ascending and descending tracks under the visibility differentiation map to model and solve the problem, so as to obtain two-dimensional deformation data that can better reflect the actual condition of the slope.

[0056] In one possible implementation, deformation calculation is performed on the two-dimensional deformation information within the visibility differentiation map to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively. This can be achieved through the following steps 1021 to 1023, which are explained in detail below.

[0057] Step 1021: Decompose the deformation of the slope along the radar line of sight in the two-dimensional deformation information into multiple deformation components.

[0058] Here, the azimuth angle of the ascending and descending orbit time-series SAR satellite imagery can be used. Angle of incidence Topographic slope Slope direction The parameters are used to decompose the slope deformation along the radar line of sight in the two-dimensional deformation information into three deformation components. The deformation components include the vertical deformation components. Slope direction deformation component and the deformation component of the ridge orientation The decomposition process can be represented by the following formula: (2) Since highway slope deformation is often caused by gravity-induced landslides along the slope surface, and it is difficult for deformation to occur along the ridgeline, it can be... The default value is 0, so the above deformation components It can be simplified to: (3) Step 1022: In the dual-track visible area of ​​the visibility differentiation map, calculate the deformation of the ascending and descending tracks in the two-dimensional deformation information based on the deformation components under the ascending and descending tracks to obtain the deformation characteristics of the target area in the vertical and slope directions, respectively.

[0059] Here, as Figure 2As shown, after decomposing the two-dimensional deformation information to obtain the LOS deformation of the ascending and descending tracks, the two-dimensional deformation calculation of the slope can be performed. First, the double-track visible area in the visibility distinction map is identified. This area can exhibit two-dimensional deformation information from both the ascending and descending track observation angles. Therefore, based on the deformation components under the ascending and descending tracks, the ascending and descending track deformation variables in the two-dimensional deformation information are calculated to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively. The ascending and descending track deformation variables in the two-dimensional deformation information include the ascending track LOS deformation and the descending track LOS deformation, denoted as... , The solution formula is as follows: (4) (5) in, The azimuth angle of the time-series SAR satellite imagery in the ascending and descending orbits. Angle of incidence For terrain slope, In the direction of the slope, This indicates LOS deformation for ascending orbitals. This indicates a LOS deformation during a drop orbit. and These are the vertical deformation component and the slope deformation component, respectively.

[0060] By solving the equations in formulas (4) and (5) respectively, the unknown can be obtained. and This allows us to determine deformation characteristics in the vertical and slope directions. Deformation characteristics include deformation rate and cumulative slope deformation value. Since time-series SAR satellite imagery in ascent and descent orbits has temporal characteristics, deformation characteristics also include cumulative deformation values ​​over a cumulative period, such as the cumulative deformation value over the most recent month.

[0061] Step 1023: In the single-track visible area of ​​the visibility differentiation map, calculate the deformation of the rising or falling track in the two-dimensional deformation information based on the deformation components under the rising or falling track to obtain the deformation characteristics of the target area in the vertical direction and the slope direction respectively.

[0062] For a single-track visible area in the visibility distinction map, select the deformation of the ascending or descending track from the ascending or descending track deformation variables, based on whether the single-track visible area represents an ascending or descending track. The solution process involves calculating the LOS deformation of the ascending or descending orbit based on the deformation components obtained from the decomposition. This yields the deformation characteristics of the target area in the vertical and slope directions, respectively. The calculation formulas are as follows: (6) (7) in, Indicates LOS deformation of the ascending orbit. Or LOS deformation due to a dropout , Indicates the angle of incidence of SAR satellite imagery. This represents the angle between the SAR satellite's incident direction and the slope direction. The unknown can be determined by solving the equations in formulas (6) and (7) above. and This allows us to determine the deformation characteristics in the vertical and slope directions. The deformation characteristics include the cumulative deformation value of the slope and the cumulative deformation value over the cumulative time period.

[0063] The invisible areas in the visibility distinction diagram cannot be determined because the LOS deformation of the ascending track cannot be determined. and LOS deformation This area falls within the satellite's observation blind zone and does not require calculation.

[0064] In this embodiment of the invention, the deformation along the radar line of sight is decomposed into deformation components in the vertical direction and the slope direction. Deformation calculation of two-dimensional deformation information is realized in the dual-track and single-track visible areas. Finally, the deformation characteristics of the target area in the slope direction and the vertical direction are extracted and used as the basis for subsequent judgment of road slope hidden danger sections.

[0065] Step 103: Determine the sections of the highway with potential slope hazards based on the deformation characteristics.

[0066] like Figure 2 As shown, after obtaining the deformation characteristics in the slope direction and the vertical direction through two-dimensional deformation calculation of the slope, intelligent screening of potential road hazards on slope sections is further carried out. First, deformation feature vectors are constructed, and then the deformation heat map of the road area is determined through kernel density analysis to realize the extraction and location of potential road hazard slope sections.

[0067] In one possible implementation, determining the potential slope hazards of the highway under test based on deformation characteristics can be achieved through the following steps 1031 to 1034, which are explained in detail below.

[0068] Step 1031: Construct a deformation feature vector based on the deformation features, calculate the kernel density corresponding to each deformation point in the target region based on the deformation feature vector, and generate the corresponding deformation heat map.

[0069] This step first determines the deformation rate along the slope direction. Cumulative deformation value Cumulative deformation value over the cumulative period and the deformation rate in the vertical direction Cumulative deformation value and cumulative deformation value over the cumulative period Then, based on these 6 features, a deformation feature vector is constructed for the deformation point, denoted as X. X can be expressed as... Next, the number of deformation points in the target region is determined, and then the kernel density corresponding to each deformation point is calculated based on the kernel density function of the deformation characteristics, denoted as f(X). The calculation formula is as follows: (8) Where K is the Gaussian kernel function, d represents the dimension of the deformation feature vector X, L represents the size range of the target region (i.e., the target region is L×L), and n represents the number of deformation points within the L×L range. Let represent the deformation eigenvector of the i-th deformation point.

[0070] The kernel density of all deformation points in the target area can be calculated using the above formula (8), thereby generating the corresponding deformation hotspot map within the target area.

[0071] Step 1032: Obtain the route vector map and slope deformation rate raster map of the highway to be tested, and determine the preset buffer area on both sides of the highway route in the route vector map. Extract the target heat map from the deformation heat map in the preset buffer area.

[0072] Here, the route vector map of the highway to be tested can be pre-drawn, while the slope deformation rate raster map is constructed using the slope deformation rate from the deformation features. First, the deformation heat map, slope deformation rate raster map, and route vector map are uniformly transformed to the same spatial reference coordinate system. Then, a spatial overlay analysis method can be used to construct a preset buffer zone on both sides of the highway route as the centerline, with a preset buffer radius. Next, the preset buffer zone is spatially intersected with the deformation heat map, and the deformation heat map regions falling within the preset buffer zone are extracted to obtain the target heat map.

[0073] Step 1033: Determine the area with a kernel density greater than the kernel density threshold in the target hotspot map as the deformation cluster area, and spatially overlay the deformation cluster area with the slope deformation rate raster map to obtain the target cluster area.

[0074] This step involves kernel density screening of the target heatmap. Since kernel density is correlated with potential slope hazards on highways, the kernel density values ​​of historical slope disasters along the highway under test can be determined. This kernel density value is then used as the kernel density threshold for historical disasters, which is used to assess deformation in the target heatmap. When the kernel density of a deformation point exceeds this threshold, it indicates a potential historical disaster hazard at that point. Therefore, areas in the target heatmap with kernel densities greater than the threshold are identified as deformation clusters. Further, these deformation clusters are overlaid pixel-by-pixel with the slope deformation rate raster map to obtain the target cluster area.

[0075] Step 1034: When the deformation hotspot level of the target cluster area exceeds the level threshold and the slope deformation rate exceeds the rate threshold, the highway route in the target cluster area is identified as the slope hazard section of the highway to be tested.

[0076] Here, slope deformation parameters such as the mean, maximum, and coefficient of variation of slope deformation rate within the target cluster area are statistically analyzed to quantify deformation intensity and non-uniformity. Furthermore, the slope deformation parameters of the target cluster area are assessed using established deformation hotspot classification rules to determine the corresponding deformation hotspot level.

[0077] When the deformation hotspot level of a target cluster exceeds a preset level threshold and the preset slope deformation rate exceeds a rate threshold, the target cluster can be identified as a high-risk hazard area. Here, the high-risk hazard area is associated with the road route of the highway to be tested. The road route located in the high-risk hazard area can be identified as the slope hazard section of the highway to be tested. The precise location and start and end mileage of the slope hazard section are output through spatial visualization, realizing the screening and location of hazard slope sections.

[0078] According to the embodiments of the present invention, based on the extracted deformation features in the vertical and slope directions, high-risk hazard areas can be accurately screened using kernel density analysis, enabling rapid screening and location of road sections with potential slope hazards, and providing accurate monitoring targets for subsequent risk prediction by UAV InSAR.

[0079] Step 104: Based on the UAV SAR imagery obtained from the road section with potential slope hazards, an interferometric phase map with deformation phase is obtained, and the deformation characteristics in the road section with potential slope hazards are inverted according to the interferometric phase map to obtain deformation time series information.

[0080] For road sections with potential hazards identified by satellite InSAR screening, drones equipped with SAR systems are used to conduct mobile, high-frequency repetitive orbital interferometry measurements to monitor and provide early warning for road sections with potential safety hazards that require real-time monitoring and handling.

[0081] like Figure 3 As shown, in the UAV monitoring phase, the first step is to design the UAV system flight path parameters, which mainly involves two aspects: first, confirming the SAR system's incident angle; and second, designing a terrain-adaptive flight path. The UAV SAR system uses side-looking imaging, and undulating terrain is prone to shadowing and overlay phenomena, significantly interfering with radar imaging quality. Therefore, the incident angle should be dynamically designed based on the slope of the potentially hazardous road section and the radar beam coverage requirements. If the maximum slope of the potentially hazardous road section is... When the drone's side-view imaging faces the slope, the incident angle must be satisfied. When facing away from the slope, the angle of incidence must be satisfied. .

[0082] In terrain-adaptive flight path design, the initial flight path and the ground swath width of the imaging strip are first determined: based on the high-precision DEM and the incident angle. The relative flight altitude H of the UAV, and the radar beamwidth. Based on the geometric relationship of SAR imaging, the horizontal ground swath width of the initial SAR imaging strip is calculated. The formula is as follows: (9) Then, based on the geometric relationship of the central projection imaging, the position of the track line corresponding to the ground strip coverage area is calculated. .

[0083] Next, the effective lateral overlap is calculated by combining the terrain slope and aspect. The formula is as follows: (10) in, This is the preset minimum lateral overlap. As a terrain adjustment factor, the imaging surface is the upslope surface. The radar beam is compressed on the upslope surface, narrowing the ground coverage. To ensure overlap, the flight tracks need to be denser. When the imaging surface is a back slope, the ground cover becomes wider, and the flight tracks can be relatively sparse. .

[0084] Furthermore, based on the effective lateral overlap, the horizontal track spacing between adjacent routes is calculated. The expression is as follows: (11) Based on this, the first route will be shifted parallel to the target. By determining the distance, the horizontal position of the second flight path can be obtained. Based on the position of the second flight path and the same SAR geometric parameters, the range of the second ground strip covered by its imaging strip can be calculated. Then repeat the above two steps to... The calculation continues as a new baseline strip, thereby iteratively generating a route network covering the entire hazardous slope section area.

[0085] Once the incident angle of the SAR system and the flight path network of the UAV are determined, UAV SAR image acquisition can be performed. The UAV, equipped with the SAR system, is set up to take pictures on the hazardous slope section. The imaging time of each SAR image is recorded during shooting, thereby obtaining multiple UAV SAR images with a temporal sequence.

[0086] Further, UAV SAR image registration is performed. Specifically, this involves using the acquired sequence of N UAV SAR images. (In chronological order), respectively with Main image , , , This forms the corresponding secondary image. Then, combining the terrain characteristics of the potentially hazardous slope section, an UAV SAR image registration method with an adaptive window shape is adopted. The window is lengthened along the contour line on the slope, and extended along the direction of the ridge or valley. In flat areas, a conventional rectangular window is used, thereby ensuring improved registration accuracy in difficult areas.

[0087] See also Figure 3 After the registration of UAV SAR images is completed, UAV SAR image interferometry is further performed to obtain an interferometric phase map with deformed phase.

[0088] In one possible implementation, an interferometric phase map with deformation phase is obtained based on UAV SAR images acquired from road sections with potential slope hazards. This can be achieved through the following steps 1041 to 1046, which are explained in detail below.

[0089] Step 1041: Based on the UAV SAR imagery obtained from the road section with potential slope hazards, an initial interferometric phase map is obtained.

[0090] Here, the primary and secondary images for image registration are first determined, and then the primary and secondary images are multiplied by conjugate to generate the initial interferometric phase map.

[0091] Step 1042: Calculate the flat-ground phase component based on the sensor spatial position coordinates of the UAV SAR image at the imaging time, and remove the flat-ground phase component from the initial interferometric phase map to obtain the target coherent phase map.

[0092] Specifically, the imaging times of the main image and the secondary image are first extracted separately. Then, the spatial position coordinates of the UAV sensors at the imaging times of the main image and the secondary image are determined separately. Next, the spatial baseline components of these two spatial position coordinates in the vertical heading direction are calculated. Combined with the imaging geometric parameters, a spatial baseline model is constructed, and the flat-ground phase component is calculated and removed. The formula is as follows: (12) in, For the wavelength of SAR system satellites, R represents the vertical baseline of the SAR system satellites, and R represents the slant range. The incident angle of the SAR system, The difference in slant distance between the main image and the secondary image.

[0093] Then, the flat-ground phase component is removed from the initial interferometric phase map to obtain the target coherent phase map.

[0094] Step 1043: Filter out high coherence points from the pixels of the target coherent phase map whose temporal coherence coefficient is greater than the coefficient threshold.

[0095] Here, for each pixel in the target coherent phase map, the temporal coherence coefficient is calculated, and a set coefficient threshold is obtained. Pixels with temporal coherence coefficients greater than the coefficient threshold are selected as high coherence points and used as control points for subsequent phase unwrapping and error correction.

[0096] Step 1044: Construct a quadratic polynomial model with the planar coordinates of the high coherence points as the independent variable and the phase value of the high coherence points in the target coherence phase diagram as the dependent variable.

[0097] Here, a model is constructed focusing on highly coherent points. The planar coordinates of these points are used as independent variables, and their phase values ​​in the target coherence phase map are used as dependent variables, resulting in a quadratic polynomial model. This quadratic polynomial model characterizes the orbital trend phase caused by residual spatial baseline errors, denoted as […]. The expression for the quadratic polynomial model is as follows: (13) Where x is the planar coordinate of the highly coherent point, and y is the phase value of the highly coherent point in the target coherent phase map. , , , , , where are the coefficients of the polynomial to be fitted.

[0098] Step 1045: Analyze the polynomial coefficients of the quadratic polynomial model and calculate the optimal estimate of the trajectory trend phase based on the polynomial coefficients.

[0099] Here, the least squares method is used to solve for the polynomial coefficients to be fitted based on actual observation data from high coherence points. Based on the solved polynomial coefficients and the aforementioned quadratic polynomial model, the orbital trend phase can be further calculated. The optimal estimate.

[0100] Step 1046: Generate an orbital trend correction surface of the same size as the initial interferometric phase map based on the optimal estimate, and remove the orbital trend correction surface from the target coherent phase map to obtain an interferometric phase map with deformed phase.

[0101] Here, based on the optimal estimate of the orbital trend phase at each high coherence point, a corresponding orbital trend correction surface can be generated. The size of this correction surface is the same as that of the initial interferometric phase map. Further, to eliminate residual spatial baseline errors, error correction is performed by removing the orbital trend correction surface from the target coherent phase map, finally yielding an interferometric phase map that retains the deformation phase.

[0102] In this embodiment of the invention, by performing UAV SAR image interferometry, the flat-ground phase component and orbital trend phase are calculated in real time and removed from the initial interferometric phase map of the SAR image. This effectively eliminates the spatial baseline error present in the image and ultimately preserves the deformation phase information, thereby improving the accuracy of deformation information extraction.

[0103] like Figure 3 As shown, after completing the UAV SAR image interferometry, the deformation characteristics of the hazardous road section are then inverted. Specifically, the deformation characteristics of the hazardous road section on the slope are inverted based on the interferometric phase map to obtain the deformation time sequence information.

[0104] Here, the weighted average of the deformation phase change rate is calculated for the interferometric phase diagram, and denoted as the deformation phase change rate. The formula is as follows: (14) Where M represents the number of interferometric phase patterns, and i represents the i-th interferometric phase pattern. This represents the deformation phase in the i-th interferometric phase diagram. This represents the time interval in the i-th interferometric phase diagram.

[0105] Furthermore, the radar wavelength of the SAR system is then utilized. The deformation phase change rate Converted to deformation rate, denoted as The formula is as follows: (15) Similarly, using the above conversion method, for the time interval of the i-th interferometric phase pattern The deformation time interval is obtained by conversion. The formula is as follows: (16) The above conversion yielded In reality, it's a deformation time series. Based on this deformation time series, by accumulating the deformation time intervals, we can obtain the deformation phase of each subsequent deformation time stamp relative to the initial deformation time stamp. Subtracting these subtractions yields the deformation phase change information of the latest deformation time stamp relative to the previous one. Therefore, by using the deformation phase of each deformation time stamp in the above deformation time series, we can obtain deformation time series information. The time span of this deformation time series information is the monitoring period of the UAV SAR system, typically set to 3 months, with each deformation time stamp generally consisting of 10 times, and the deformation time intervals generally around 9 days.

[0106] Step 105: Perform slope risk prediction based on deformation time series information to obtain slope monitoring results for the highway under test.

[0107] like Figure 3 As shown, for each road section with potential slope hazards, based on the deformation time series information obtained by inverting the above deformation characteristics, long-term deformation characteristics and short-term deformation characteristics can be extracted. Then, on this basis, a slope risk classification and early warning model can be constructed to provide early warning of slope risks for road sections with potential slope hazards, thereby realizing slope risk monitoring of the highway under test.

[0108] In one possible implementation, slope risk prediction is performed on deformation time series information to obtain slope monitoring results of the highway under test. This can be achieved through the following steps 1051 to 1055, which are explained in detail below.

[0109] Step 1051: Extract long-term and short-term deformation features from road sections with potential slope hazards based on deformation time series information.

[0110] Among these, long-term deformation characteristics can cover the entire monitoring period of the UAV SAR system, that is, the time span covering deformation time-series information (from the first deformation timestamp to the last deformation timestamp). Specifically, this includes the average deformation rate. Average cumulative deformation value Deformation space heterogeneity index .in It can be calculated based on the deformation change and time span in the deformation time series information, representing the average deformation rate of the slope hazard section during this monitoring period. It is the average value of the total deformation at each deformation point in the road section with potential slope hazards during the monitoring period. It is obtained by first calculating the ratio of the standard deviation to the average value of the deformation rate of each deformation point in the slope hazard section within the monitoring period, and then calculating the average value.

[0111] Short-term deformation characteristics typically only cover a portion of the monitoring period monitored by the UAV SAR system, that is, a portion of the time span covering deformation time-series information. For example, this might include the average deformation value of a slope section with potential hazards, based on the latest deformation timestamp. Mean deformation acceleration Deformation rate angle Spatial coherence decline index . It is the average value of the deformation of each deformation point in the road section with potential slope hazards within the latest deformation timestamp; It is the ratio of the difference in the average deformation rate of the road section with potential slope hazards to the time interval between the two most recent adjacent deformation timestamps; It is calculated using the arctangent function as the ratio of the average deformation to the time interval between the two most recent adjacent deformation timestamps. It is the ratio of the average coherence coefficient of the latest deformation timestamp slope hazard section to the average coherence coefficient of the historical reference period (stable state of highway slope).

[0112] After extracting the three long-term deformation features and four short-term deformation features respectively, these seven deformation features are then uniformly normalized to obtain the normalized deformation features, denoted as . .

[0113] Step 1052: When the warning level of the slope hazard section is lower than the preset standard warning level and the duration of the warning level being lower than the preset standard warning level is greater than the time threshold, assign a first weight to each long-term deformation feature and set a second weight for each short-term deformation feature.

[0114] The slope risk classification and early warning model is used to calculate the risk score of highway slopes by combining deformation characteristics and set feature weights. The feature weights need to be set according to certain conditions. Generally, after identifying a road section with potential slope hazards, an early warning level is assessed for that section. The higher the early warning level, the more serious the slope hazard. If the early warning level exceeds the standard early warning level, the disaster risk of the slope hazard is higher; conversely, if it does not exceed the standard level, the disaster risk is lower.

[0115] When the warning level of a highway slope is lower than the preset standard warning level, and the duration of the warning level being lower than the preset standard warning level is greater than the time threshold (e.g., a time span greater than 5 deformation timestamps), it indicates that the road section with potential slope hazards is in a low disaster risk state for a long time. Therefore, the long-term deformation characteristics that reflect the long-term stability of the slope are given a first weight that is higher than the second weight, while the short-term deformation characteristics that reflect the short-term changes of the slope are given a second weight that is lower than the first weight, i.e., the first weight is greater than the second weight.

[0116] Step 1053: When the warning level of the road section with potential slope hazards exceeds the preset standard warning level, assign a first weight to each long-term deformation feature and set a second weight for each short-term deformation feature.

[0117] When the warning level of a road section with potential slope hazards exceeds the preset standard warning level, it indicates that the road section with potential slope hazards is in a state of high disaster risk. Therefore, a second weight with a higher weight than the first weight is set for the short-term deformation characteristics that reflect the short-term changes of the slope, while a first weight with a lower weight than the second weight is assigned to the long-term deformation characteristics that reflect the long-term stability of the slope. That is, the second weight is greater than the first weight.

[0118] Step 1054: Calculate the highway slope risk score based on short-term deformation characteristics, long-term deformation characteristics, first weight, and second weight.

[0119] Here, short-term deformation characteristics, long-term deformation characteristics, and the set first and second weights are all input into the slope risk classification and early warning model for calculation to obtain the highway slope risk score, denoted as F. The calculation formula of the slope risk classification and early warning model is as follows: (17) in, This represents the normalized value of the i-th deformation feature, i.e., the normalized value of the long-term deformation feature or the short-term deformation feature. This represents the weight set for the i-th deformation feature, i.e., the first weight or the second weight.

[0120] Furthermore, the weight assigned to each short-term or long-term deformation feature is negatively correlated with the feature correlation. Here, in the slope risk classification and early warning model, weights are adjusted based on deformation features, specifically by calculating the feature correlation between these features, such as feature similarity. The higher the feature similarity between two deformation features, the higher the feature correlation, indicating that the two deformation features are more similar, and therefore the smaller the feature weight (i.e., the first weight or the second weight) assigned to these two deformation features. Conversely, the lower the feature similarity between two deformation features, the lower the feature correlation, indicating that the two deformation features are more different, and therefore the larger the feature weight (i.e., the first weight or the second weight) assigned to these two deformation features.

[0121] Step 1055: Determine the risk level of the highway slope based on the preset highway slope risk level threshold, and use the determined highway slope risk level as the slope monitoring result of the highway to be tested.

[0122] In this embodiment of the invention, the risk score range for risky slopes is divided according to a preset risk level threshold for highway slopes, with each range representing a risk warning level. For example, the risk level thresholds for highway slopes are as follows: , , This is thus divided into four value ranges, corresponding to four risk warning levels: blue, yellow, orange, and red. The specific rules for this division will be explained below.

[0123] Specifically, if the highway slope risk score F is less than If the deformation characteristics of the road section with potential slope hazards remain stable and there are no obvious signs of slope instability, then the risk warning level is determined to be a blue warning.

[0124] And if F is greater than and less than or deformation rate angle Greater than or average cumulative deformation value Greater than or average deformation rate Greater than If the risk level is yellow, it indicates that the deformation characteristics of the road section with potential slope hazards have already shown significant fluctuations, and there is already a slight slope instability.

[0125] And if F is greater than and less than or deformation rate angle Greater than or average cumulative deformation value Greater than or average deformation rate Greater than If the risk level is positive, it indicates that the deformation characteristics of the road section with potential slope hazards have changed significantly, indicating that the slope is unstable and there is a possibility of slope disaster. Therefore, the risk warning level is determined to be an orange warning.

[0126] And if F is greater than or deformation rate angle Greater than or average cumulative deformation value Greater than or average deformation rate Greater than If the deformation characteristics of the road section with potential slope hazards have shown abnormal fluctuations, it indicates that slope instability may have occurred, or even slope disaster may have occurred. In this case, the risk warning level is determined to be a red warning.

[0127] When classifying risk warning levels, the calculated risk score of the highway slope is determined according to the above-mentioned risk warning level classification rules, thereby determining its corresponding risk warning level, which serves as the slope monitoring result for road sections with potential slope hazards.

[0128] For each identified road section with potential slope hazards, the road slope risk score is calculated using the same method described above. Then, based on the risk warning level classification rules, the risk warning level corresponding to each road section with potential slope hazards is determined. The determined road slope risk level is then used as the slope monitoring result of the road to be monitored, thereby realizing the slope monitoring of the target area of ​​the road to be monitored.

[0129] Furthermore, by setting corresponding warning signs for the classified risk warning levels, the risk warning level of each potential slope section in the target area of ​​the highway under test can be visualized, clearly displaying the slope monitoring results of the highway under test.

[0130] This invention proposes a slope risk classification and early warning model that can dynamically adjust weights by combining temporal deformation characteristics. Finally, the corresponding slope risk level is determined based on the highway slope risk score calculated by the model. This achieves intelligent and hierarchical early warning of slope risks, significantly improving the timeliness and accuracy of highway slope safety monitoring.

[0131] In summary, this invention utilizes time-series SAR satellite imagery of the target area of ​​the highway under test to acquire deformation features in the vertical and slope directions, thereby initially locating potential slope hazards on the highway. Furthermore, UAV SAR imagery is used to focus on these hazardous slope sections, extracting time-series deformation features for slope risk prediction, thus obtaining the slope monitoring results for the highway under test. This collaborative monitoring method combining satellite InSAR and UAV InSAR not only leverages the advantages of both InSAR platforms and overcomes the limitations of single-platform InSAR in highway monitoring, but also achieves seamless integration from large-scale highway network screening to precise local monitoring. By analyzing two-dimensional deformation information along the radar line of sight to extract deformation features in the vertical and slope directions of the target area, it comprehensively and realistically reflects the movement characteristics of highway slopes in different directions, overcoming the deficiency of insufficient one-dimensional deformation information and improving the accuracy of slope monitoring in complex highway terrain.

[0132] The following section details the highway slope monitoring device based on multi-platform InSAR collaboration provided by this invention.

[0133] Figure 4 This is a schematic diagram of the structure of the highway slope monitoring device based on multi-platform InSAR collaboration provided by the present invention, as shown below. Figure 4 As shown, the highway slope monitoring device based on multi-platform InSAR collaboration specifically includes: a two-dimensional deformation calculation module 401, a deformation feature extraction module 402, a slope hazard determination module 403, a deformation feature inversion module 404, and a slope risk prediction module 405.

[0134] Specifically, the two-dimensional deformation calculation module 401 is used to extract surface deformation information under ascending and descending orbits from the ascending and descending orbit time-series SAR satellite images of the target area corresponding to the highway under test, and to perform surface deformation phase calculation to obtain two-dimensional deformation information along the radar line of sight under the ascending and descending orbit time-series SAR satellite images; the deformation feature extraction module 402 is used to identify the visible area of ​​the ascending and descending orbit time-series SAR satellite images to obtain a visibility distinction map of the highway under test, and to perform deformation calculation on the two-dimensional deformation information within the visibility distinction map to obtain the target area. The system includes: deformation characteristics in the vertical and slope directions; a slope hazard determination module 403, used to determine the slope hazard sections of the highway under test based on the deformation characteristics; a deformation characteristic inversion module 404, used to obtain an interferometric phase map with deformation phase based on UAV SAR images obtained from the slope hazard sections, and to invert the deformation characteristics in the slope hazard sections based on the interferometric phase map to obtain deformation time series information; and a slope risk prediction module 405, used to predict the slope risk based on the deformation time series information to obtain the slope monitoring results of the highway under test.

[0135] The highway slope monitoring device based on multi-platform InSAR collaboration provided in the above embodiments can realize the technical solutions described in the above embodiments of the highway slope monitoring method based on multi-platform InSAR collaboration. The specific implementation principles of each module or unit can be found in the corresponding content in the above embodiments of the highway slope monitoring method based on multi-platform InSAR collaboration, and their technical effects can also be referred to each other, which will not be repeated here.

[0136] like Figure 5 As shown, the present invention also provides an electronic device 500. The electronic device 500 includes a processor 501, a memory 502, and a display 503. Figure 5 Only some components of the electronic device 500 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.

[0137] In some embodiments, memory 502 may be an internal storage unit of electronic device 500, such as a hard disk or memory of electronic device 500. In other embodiments, memory 502 may also be an external storage device of electronic device 500, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 500.

[0138] Furthermore, the memory 502 may include both internal storage units of the electronic device 500 and external storage devices. The memory 502 is used to store application software and various types of data installed on the electronic device 500.

[0139] In some embodiments, processor 501 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 502 or process data, such as the multi-platform InSAR collaborative highway slope monitoring method of the present invention.

[0140] In some embodiments, display 503 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 503 is used to display information from electronic device 500 and to display a visual user interface. Components 501-503 of electronic device 500 communicate with each other via a system bus.

[0141] In some embodiments of the present invention, when the processor 501 executes the computer program in the memory 502, the following steps can be implemented: extracting surface deformation information under the rising and falling orbit from the rising and falling orbit time-series SAR satellite image of the target area corresponding to the highway to be tested, and performing surface deformation phase calculation to obtain two-dimensional deformation information along the radar line of sight under the rising and falling orbit time-series SAR satellite image; performing visible area identification on the rising and falling orbit time-series SAR satellite image to obtain a visibility distinction map of the highway to be tested, and performing deformation calculation on the two-dimensional deformation information within the visibility distinction map to obtain the deformation characteristics of the target area in the vertical direction and the slope direction respectively; determining the slope hazard sections of the highway to be tested based on the deformation characteristics; obtaining an interferometric phase map with deformation phase based on the UAV SAR image obtained from the slope hazard sections, and performing inversion on the deformation characteristics in the slope hazard sections based on the interferometric phase map to obtain deformation time-series information; performing slope risk prediction on the deformation time-series information to obtain the slope monitoring results of the highway to be tested.

[0142] It should be understood that when the processor 501 executes the computer program in the memory 502, in addition to the functions described above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.

[0143] Furthermore, the embodiments of the present invention do not specifically limit the type of electronic device 500 mentioned. Electronic device 500 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, electronic device 500 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0144] Furthermore, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, this computer program performs the aforementioned method for monitoring highway slopes based on multi-platform InSAR collaboration. This method includes: extracting surface deformation information under the rising and falling orbits from the rising and falling orbit time-series SAR satellite images of the target area corresponding to the highway to be monitored, and performing surface deformation phase calculation to obtain two-dimensional deformation information along the radar line-of-sight direction under the rising and falling orbit time-series SAR satellite images; and performing visible area identification on the rising and falling orbit time-series SAR satellite images to obtain... A visibility map of the highway to be tested is obtained. Within the visibility map, deformation calculation is performed on the two-dimensional deformation information to obtain the deformation characteristics of the target area in the vertical and slope directions, respectively. Based on the deformation characteristics, the slope hazard sections of the highway to be tested are identified. Based on UAV SAR images obtained from the slope hazard sections, an interferometric phase map with deformation phase is obtained. The deformation characteristics in the slope hazard sections are inverted based on the interferometric phase map to obtain deformation time series information. Slope risk prediction is performed on the deformation time series information to obtain the slope monitoring results of the highway to be tested.

[0145] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0146] The above provides a detailed description of the highway slope monitoring method and device based on multi-platform InSAR collaboration provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A highway slope monitoring method based on multi-platform InSAR cooperation, characterized in that, include: The surface deformation information under the rising and falling orbit is extracted from the rising and falling orbit time-series SAR satellite image of the target area corresponding to the highway to be tested, and the surface deformation phase is calculated to obtain the two-dimensional deformation information along the radar line of sight under the rising and falling orbit time-series SAR satellite image. The visible area is identified by the time-series SAR satellite imagery of the ascending and descending orbits to obtain a visibility distinction map of the highway to be measured. Within the visibility distinction map, the deformation information of the two-dimensional deformation is calculated to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively. Based on the deformation characteristics, identify the sections of the highway with potential slope hazards. Based on UAV SAR images obtained from the road section with potential slope hazards, an interferometric phase map with deformation phase is obtained, and the deformation characteristics in the road section with potential slope hazards are inverted according to the interferometric phase map to obtain deformation time series information. The deformation time series information is used to predict slope risk, and the slope monitoring results of the highway to be tested are obtained.

2. The highway slope monitoring method based on multi-platform InSAR cooperation according to claim 1, characterized in that, The surface deformation information includes permanent scatterer points and distributed scatterer points in the target area. The surface deformation phase calculation process for the permanent scatterer points and the distributed scatterer points includes: The permanent scatterer points and the distributed scatterer points are used as nodes to construct a Delaunay triangulation, and differential interference phase calculation is performed on the edges in the Delaunay triangulation to obtain the absolute phase of the nodes. Obtain digital elevation model data and GACOS atmospheric correction data for the target area; Based on the terrain interferometric phase simulated by the digital elevation model data, the absolute phase is unwrapped, and atmospheric delay phase removal is performed on the unwrapped continuous phase field based on atmospheric correction data to obtain the deformation phase of the target area. Spatiotemporal parameters are inverted on the deformation phase to obtain two-dimensional deformation information along the radar line of sight in the ascending and descending orbit time-series SAR satellite image.

3. The highway slope monitoring method based on multi-platform InSAR collaboration according to claim 1, characterized in that, The ascending and descending orbit time-series SAR satellite imagery includes both ascending and descending orbit time-series SAR satellite imagery of the target area. The process of identifying the visible area from the ascending and descending orbit time-series SAR satellite imagery to obtain a visibility differentiation map of the highway under test includes: Based on the incident angle, terrain slope, and slope direction of the ascending and descending orbit time-series SAR satellite images, the R-index is calculated for ascending and descending orbits respectively. The visible and invisible areas of the target region are determined based on the R index; The target area that is divided into visible areas in the time-series SAR satellite imagery of the ascending and descending orbits is identified as the dual-track visible area of ​​the highway to be measured. The target area that is divided into a visible area in the ascending-descending orbit time-series SAR satellite image or the descending orbit time-series SAR satellite image is determined as the single-track visible area of ​​the highway to be measured; The target areas that are all classified as out-of-view areas in the time-series SAR satellite images of the ascending and descending orbits are identified as the out-of-view areas of the highway to be measured. A visibility differentiation map of the highway under test is constructed based on the dual-track visible area, the single-track visible area, and the invisible area.

4. The highway slope monitoring method based on multi-platform InSAR collaboration according to claim 1, characterized in that, Within the visibility differentiation map, deformation calculation is performed on the two-dimensional deformation information to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively, including: The deformation of the slope along the radar line of sight in the two-dimensional deformation information is decomposed into multiple deformation components, including vertical deformation components, slope surface deformation components, and ridge orientation deformation components. In the dual-track visible area of ​​the visibility differentiation map, the deformation of the lifting and lowering tracks in the two-dimensional deformation information is calculated based on the deformation components under the lifting and lowering tracks to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively. In the single-track visible area of ​​the visibility differentiation map, the deformation of the rising or falling track in the two-dimensional deformation information is calculated based on the deformation component under the rising or falling track to obtain the deformation characteristics of the target area in the vertical direction and the slope direction, respectively.

5. The highway slope monitoring method based on multi-platform InSAR collaboration according to claim 1, characterized in that, The process of determining the slope hazard sections of the highway to be tested based on the deformation characteristics includes: Based on the deformation features, a deformation feature vector is constructed, and based on the deformation feature vector, the kernel density corresponding to each deformation point in the target region is calculated, and a corresponding deformation heat map is generated. Obtain the route vector map and slope deformation rate raster map of the highway to be tested, and determine the preset buffer area on both sides of the highway route in the route vector map. Extract the target heat map from the deformation heat map in the preset buffer area. The region with a kernel density greater than the kernel density threshold in the target hotspot map is identified as a deformation clustering region, and the deformation clustering region is spatially superimposed with the slope deformation rate raster map to obtain the target clustering region. When the deformation hotspot level of the target cluster area exceeds the level threshold and the slope deformation rate exceeds the rate threshold, the road route in the target cluster area is identified as a section of the road with potential slope hazards on the road to be tested.

6. The highway slope monitoring method based on multi-platform InSAR collaboration according to claim 1, characterized in that, The process of obtaining an interferometric phase map with deformation phase based on UAV SAR imagery acquired from the road section with potential slope hazards includes: An initial interferometric phase map was obtained based on UAV SAR images acquired from the road section with potential slope hazards. Based on the sensor spatial position coordinates of the UAV SAR image at the imaging time, the flat-ground phase component is calculated, and the flat-ground phase component is removed from the initial interferometric phase map to obtain the target coherent phase map; High coherence points with temporal coherence coefficients greater than a coefficient threshold are selected from the pixels of the target coherent phase map; Using the planar coordinates of the highly coherent points as independent variables and the phase value of the highly coherent points in the target coherent phase diagram as dependent variables, a quadratic polynomial model is constructed. The quadratic polynomial model characterizes the orbital trend phase caused by residual spatial baseline errors. The polynomial coefficients of the quadratic polynomial model are analyzed, and the optimal estimate of the trajectory trend phase is calculated based on the polynomial coefficients. Based on the optimal estimate, an orbital trend correction surface of the same size as the initial interferometric phase map is generated, and the orbital trend correction surface is removed from the target coherent phase map to obtain an interferometric phase map with deformed phase.

7. The highway slope monitoring method based on multi-platform InSAR collaboration according to claim 1, characterized in that, The process of performing slope risk prediction on the deformation time series information to obtain slope monitoring results for the highway under test includes: Based on the deformation time series information, long-term and short-term deformation characteristics of the slope hazard section are extracted; When the warning level of the slope hazard section is lower than the preset standard warning level and the duration of the warning level being lower than the preset standard warning level is greater than a time threshold, a first weight is assigned to each long-term deformation feature, and a second weight is set for each short-term deformation feature, wherein the first weight is greater than the second weight. When the warning level of the slope hazard section exceeds the preset standard warning level, a first weight is assigned to each long-term deformation feature, and a second weight is set for each short-term deformation feature, wherein the second weight is greater than the first weight. The highway slope risk score is calculated based on the short-term deformation feature, the long-term deformation feature, the first weight, and the second weight, wherein the weight assigned to each short-term deformation feature or the long-term deformation feature is negatively correlated with the feature correlation. The risk level of the highway slope is determined by the preset highway slope risk level threshold, and the determined highway slope risk level is used as the slope monitoring result of the highway to be tested.

8. A highway slope monitoring device based on multi-platform InSAR collaboration, characterized in that, include: The two-dimensional deformation calculation module is used to extract the surface deformation information under the ascending and descending orbits of the time-series SAR satellite images of the target area corresponding to the highway under test, and to perform surface deformation phase calculation to obtain the two-dimensional deformation information along the radar line of sight under the time-series SAR satellite images of the ascending and descending orbits. The deformation feature extraction module is used to identify the visible area of ​​the ascending and descending orbit time-series SAR satellite image to obtain a visibility distinction map of the highway to be tested. Within the visibility distinction map, the two-dimensional deformation information is subjected to deformation calculation to obtain the deformation features of the target area in the vertical direction and the slope direction, respectively. The slope hazard identification module is used to identify the slope hazard sections of the highway to be tested based on the deformation characteristics. The deformation feature inversion module is used to obtain an interferometric phase map with deformation phase based on UAV SAR images acquired from the road section with potential slope hazards, and to invert the deformation features in the road section with potential slope hazards based on the interferometric phase map to obtain deformation time series information. The slope risk prediction module is used to predict slope risk based on the deformation time series information and obtain the slope monitoring results of the highway to be tested.

9. An electronic device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps of the highway slope monitoring method based on multi-platform InSAR collaboration as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the highway slope monitoring method based on multi-platform InSAR collaboration as described in any one of claims 1 to 7.

Images