Landslide geological disaster intelligent monitoring method based on airborne radar and oblique photography

Abstract

The invention discloses a landslide geological disaster intelligent monitoring method based on airborne radar and oblique photography, and the method comprises the steps: obtaining an orthoimage and a digital surface model through the oblique photography of an unmanned plane, and recognizing a landslide range and a ground surface crack; utilizing airborne radar scanning to construct a landslide digital elevation model; determining the thickness of a soil layer in combination with three-dimensional scanning of an exploratory well / excavation face; multi-source data are fused to calculate stability parameters, and intelligent early warning of landslide disasters is achieved; the precision, efficiency and automation level of landslide monitoring are improved.

Description

Technical Field

[0001] This invention relates to the field of geological disaster monitoring technology, and more specifically, to an intelligent monitoring method for landslide geological disasters based on airborne radar and oblique photography. Background Technology

[0002] Traditional landslide monitoring methods rely heavily on manual inspections, single-point sensors, or single remote sensing technologies, resulting in low efficiency, narrow coverage, and insufficient accuracy. Oblique photography can acquire high-resolution surface images, but it struggles to penetrate vegetation or surface cover. Airborne radar can acquire three-dimensional electrical clouds of the surface, but its use alone cannot effectively identify surface cracks and soil structure. Existing technologies lack a comprehensive monitoring system that integrates multi-source data fusion and intelligent analysis, leading to insufficient accuracy and timeliness in landslide early warning systems. Summary of the Invention

[0003] In order to solve at least one of the above-mentioned technical problems, the purpose of this invention is to provide an intelligent monitoring method for landslide geological disasters based on airborne radar and oblique photography, which can realize intelligent early warning of landslide disasters and improve the accuracy, efficiency and automation level of landslide monitoring.

[0004] This invention provides an intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography, including: By acquiring images of the disaster area through oblique photography by drones, constructing orthophoto images and digital surface models, and intelligently identifying the landslide impact range and surface crack distribution map based on orthophoto images; Airborne radar is used to perform three-dimensional scanning of the landslide impact area to obtain point cloud data of the landslide area, and a digital elevation model of the landslide is constructed by combining the digital surface model. Obtain 3D scanning data of wells / excavation faces within the landslide area, and construct digital elevation models of wells / excavation faces by combining them with the landslide digital elevation model; The crack parameters are determined by GNSS testing of landslide cracks and landslide profile mapping. The crack parameters include at least crack length, width, and orientation. Extract the thickness of different soil types from the digital elevation model of the well / excavation face; conduct a comprehensive analysis of the thickness and crack parameters of different soil types to assess the stability parameters of the current landslide geology; If the current stability parameters of the landslide geology are less than the preset first stability parameter threshold, an alarm message will be triggered.

[0005] In this solution, the steps for acquiring disaster area images through UAV oblique photography and constructing a digital surface model specifically include: Using a drone equipped with a five-lens camera to acquire multi-view image sets Each image i contains interior and exterior orientation elements: coordinates of the camera center. and rotation matrix ; Connection points between images are established by matching preset feature points, and all parameters are optimized using bundle adjustment for any image point. Its corresponding object point They satisfy the collinear equations: ,in Here, f is the scaling factor, and f is the focal length. This represents a 3x3 rotation matrix. The objective function for adjustment optimization is: (The vector represents the three-dimensional coordinates of the camera center.) ; Based on minimizing the energy function, a disparity map is determined, and a dense point cloud is generated from the photographic set. ; The dense point cloud is interpolated onto a predefined rule network to obtain a digital surface model.

[0006] In this solution, the steps for constructing the digital elevation model of the landslide specifically include: Airborne LiDAR systems measure laser round-trip time. Then, point cloud data is obtained by combining POS data. ,in ,in , This refers to the laser emission angle; This represents the three-dimensional coordinates of the GNSS antenna phase center, where R represents the laser ranging value. This represents the attitude rotation matrix provided by the inertial measurement unit. radar point cloud Point clouds in digital surface models Registration yields the transformation matrix T; Ground points are separated based on a preset cloth simulation filtering algorithm, with a preset point cloud height of Z and a simulated cloth node height of [missing value]. Solving iteratively , obtain ground point clouds ;in Indicates terrain elevation. Indicates the maximum elevation difference threshold; Ground point clouds The difference is a grid DEM, which yields the landslide digital elevation model. ,in Represents grid points; This represents the elevation value of the l-th known ground point. represents the kriging weight of the l-th known point to the current grid point, and L represents the number of known ground points used for interpolation; This indicates the digital elevation model of the landslide at grid points. The elevation value at that location.

[0007] In this scheme, the step of acquiring three-dimensional scanning data of the well / excavation face within the landslide area and constructing a digital elevation model of the well / excavation face in conjunction with the landslide digital elevation model specifically includes: Based on the 3D scanning data of the wells / excavation faces within the landslide area, the point cloud of the wells / excavation faces was determined. ; Extract the boundary point cloud B from the DEM to determine the wellhead / excavation face, and then combine the point cloud of the wellhead / excavation face. Projecting along the well wall direction onto the profile plane yields a two-dimensional elevation model of the profile. ; Projected coordinates are Where v and w are the basis vectors of the profile plane. The origin; By combining the profile elevation model with the landslide digital elevation model, a digital elevation model of the well / excavation face is obtained. Its formula is .

[0008] In this scheme, the step of extracting the thickness of different soil types in the digital elevation model of the well / excavation face specifically includes: Extract the feature vector of each point cloud in the digital elevation model of the well / excavation face; The feature vectors of the point cloud are divided according to preset rules to obtain soil layer categories; The point cloud in the digital elevation model of the well / excavation face is divided according to different soil layer categories, and a set of depth values ​​of all points in the corresponding layer is constructed. ; Let the soil layer thickness be H. If the well / excavation face is vertical, the formula is: ,in This represents the depth value of point a in soil layer A; This indicates the thickness of soil layer A; If the well / excavation face is not vertical, then extract the dip angle of the well / excavation face. Its formula is .

[0009] In this plan, the formula for comprehensively analyzing the thickness and crack parameters of different soil types to evaluate the stability parameters of the current landslide geology is as follows: For each exploration well / excavation face, match the cracks and corresponding crack parameters within the specified area; The well / excavation face is numbered 'b', and the stability parameter of the landslide where the well / excavation face is located is set as follows: Its formula is ,in Let N represent the cohesion weighting coefficient of soil layer A, and let N represent the set of all soil layer types. This indicates the length of crack J within the defined area. This indicates the width of crack J within the defined area; Let J be the angle between the direction of crack J's extension and the slope, and M be the set of cracks within the defined area. After traversing all wells / excavation faces and numbering them, the set of stability parameters for landslides in all grids containing wells / excavation faces is obtained; The stability parameters of the landslide are obtained by averaging the values ​​of the stability parameters of the grid containing all exploration wells / excavation faces.

[0010] This plan also includes: Compare any stability parameter in the stability parameter set with a preset first stability parameter threshold. If the stability parameter is less than the preset first stability parameter threshold, then take the grid where the corresponding stability parameter is located as the benchmark and extract the stability parameters of the landslide in the adjacent grid. The stability parameters of landslides in adjacent grids are compared and analyzed one by one with the preset second stability parameter threshold. If the stability parameters of landslides in adjacent grids are less than the preset second stability parameter threshold, a local warning message is triggered with the reference grid as the center and adjacent grids. The preset second stability parameter threshold is greater than the preset first stability parameter threshold.

[0011] This plan also includes: Based on the landslide's impact range, determine the stability parameters of the landslide where the exploration well / excavation face is located within the grid containing the outer edge of the landslide; If the stability parameter of the landslide where the well / excavation face is located within the grid of the outer edge of the landslide is greater than the preset first stability parameter threshold and less than the preset second stability parameter threshold, then the information of the well / excavation face will be added. Based on the addition of well / excavation face information, the outer edge of the landslide is expanded outward by a preset dimension to obtain the outer edge line; Determine the outer grid area based on the outer edge line and grid lines; At least one additional well / excavation face is added within the expanded grid area to recalculate the stability parameters of the current landslide geology.

[0012] In this scheme, after obtaining the stability parameters of the landslide where the exploratory well / excavation face is located, the following are also included: Obtain the depth value h(t) of the exploration well / excavation face at the current time node; Obtain the depth value h(t-1) of the exploration well / excavation face at the previous time node; The depth difference is obtained by subtracting the depth of the well / excavation face at the previous time point from the depth of the well / excavation face at the current time point. Based on the preset depth range into which the depth difference falls, determine the stability parameter revision coefficient of the landslide where the corresponding exploration well / excavation face is located; Multiply the stability parameter of the landslide where the well / excavation face is located by the corresponding stability parameter revision factor to obtain the revised stability parameter of the landslide where the current well / excavation face is located.

[0013] This plan also includes: Divide the depth difference by the time difference between two adjacent time points to obtain the depth backfill rate; If the depth backfill rate is greater than the preset depth backfill rate threshold, a secondary landslide warning message is generated within the grid where the corresponding well / excavation face is located. After traversing all exploration wells / excavation faces, determine the number of grids that triggered secondary landslide warning messages; If the number of grids that trigger secondary landslide warnings exceeds a preset threshold, secondary landslide warnings will be triggered for the entire monitored area.

[0014] One or more technical solutions proposed in this application have at least the following technical effects: By acquiring orthophotos and digital surface models through oblique photography by UAVs, the extent of landslides and surface cracks can be identified; a digital elevation model of the landslide can be constructed using airborne radar scanning; the thickness of the soil layer can be determined by combining three-dimensional scanning of the well / excavation face; and by fusing multi-source data to calculate stability parameters, intelligent early warning of landslide disasters can be achieved; this invention improves the accuracy, efficiency and automation level of landslide monitoring. Attached Figure Description

[0015] Figure 1 The flowchart of the intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography of the present invention is shown; Figure 2 A schematic diagram of the landslide digital elevation model of the present invention is shown; Figure 3 A schematic diagram of the landslide grid of the present invention is shown. Detailed Implementation

[0016] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0017] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0018] Figure 1 The flowchart of the intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography of the present invention is shown.

[0019] like Figure 1 As shown, this invention discloses a flowchart of an intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography, including: S101 uses oblique photography of UAVs to acquire images of the disaster area, constructs orthophoto images and digital surface models, and extracts the landslide impact range and surface crack distribution map based on intelligent recognition of orthophoto images. S102, Perform airborne radar three-dimensional scanning on the landslide impact area to obtain point cloud data of the landslide area, and construct a landslide digital elevation model in combination with the digital surface model; S103, acquire three-dimensional scanning data of the well / excavation face in the landslide area, and construct a digital elevation model of the well / excavation face by combining it with the landslide digital elevation model; S104. Determine crack parameters by GNSS testing of landslide cracks and landslide profile mapping. The crack parameters include at least crack length, width and orientation. S105, Extract the thickness of different soil types from the digital elevation model of the well / excavation face; Conduct a comprehensive analysis of the thickness and crack parameters of different soil types to assess the stability parameters of the current landslide geology; S106, If the current stability parameters of the landslide geology are less than the preset first stability parameter threshold, an alarm message will be triggered.

[0020] According to an embodiment of the present invention, the landslide impact range is first efficiently delineated by oblique photography, then the landslide impact area is precisely scanned by guided radar, and then GNSS and well / excavation face scanning are used to obtain accurate parameters for key parts. Finally, all parameters are comprehensively evaluated to obtain the current landslide geological stability parameters.

[0021] Specifically, following a pre-set drone flight path, operators used drones to conduct five-lens oblique photography of the landslide area, acquiring images with an overlap rate exceeding 80%. These images were then processed using software such as ContextCapture to generate a 5-cm resolution orthophoto map (DOM) and a digital surface model (DSM), determining the landslide's impact area. Next, drones equipped with InSAR radar were deployed to scan the landslide's impact area, acquiring high-density point clouds. In CloudCompare software, the radar point clouds were registered with the DSM, and a cloth simulation filtering (CSF) algorithm was used to filter out tree and building point clouds, resulting in a clean "landslide digital elevation model (DEM)," as shown below. Figure 2As shown, it reflects the true surface elevation. Under safe conditions, geologists set up at least one well / excavation trench within the gridded area to reveal the soil structure. Based on the point cloud reflection intensity (e.g., weak reflection for clay, strong reflection for gravel) and color (manually labeled different soil layer colors), a region growing segmentation algorithm is used to automatically divide the profile point cloud into soil layer types such as "plain fill layer" and "silty clay layer," and the vertical distribution range of each cluster is statistically analyzed. The lidar used is a D-LiDAR2200 model, whose point cloud data accuracy meets the requirements of ±5cm@100m horizontally, ±5cm@100m vertically, and a ground resolution of 3cm@150m.

[0022] According to an embodiment of the present invention, the step of acquiring disaster area images through UAV oblique photography and constructing a digital surface model specifically includes: Using a drone equipped with a five-lens camera to acquire multi-view image sets Each image i contains interior and exterior orientation elements: coordinates of the camera center. and rotation matrix ; Connection points between images are established by matching preset feature points, and all parameters are optimized using bundle adjustment for any image point. Its corresponding object point They satisfy the collinear equations: ,in Here, f is the scaling factor, and f is the focal length. This represents a 3x3 rotation matrix. The objective function for adjustment optimization is: (The vector represents the three-dimensional coordinates of the camera center.) ; Based on minimizing the energy function, a disparity map is determined, and a dense point cloud is generated from the photographic set. ; The dense point cloud is interpolated onto a predefined rule network to obtain a digital surface model.

[0023] It should be noted that the numerical surface model is set as Its formula is , ,in Indicates the digital surface model at grid points The elevation values ​​at the location, where i and j represent the row and column indices of the DEM grid. The search domain is defined as the set of original point cloud data that participates in the current grid point difference calculation, where k represents the search domain. The index of a certain original point within, This represents the weight of the k-th original point, which is inversely proportional to the squared distance. This represents the elevation value of the k-th original point. Represents grid points The horizontal distance to k original points; Represents extremely small positive numbers, such as This is to prevent the denominator from being zero and to ensure numerical stability; Indicates image index, ; Indicates the object point index. ; Indicates the first The object point at the th The actual measured coordinates of the image points on the image, in pixels; This represents the image point coordinates calculated based on the current exterior orientation elements and the object point coordinates.

[0024] According to an embodiment of the present invention, the step of constructing a digital elevation model of a landslide specifically includes: Airborne LiDAR systems measure laser round-trip time. Then, point cloud data is obtained by combining POS data. ,in ,in , This refers to the laser emission angle; This represents the three-dimensional coordinates of the GNSS antenna phase center, where R represents the laser ranging value. This represents the attitude rotation matrix provided by the inertial measurement unit. radar point cloud Point clouds in digital surface models Registration yields the transformation matrix T; Ground points are separated based on a preset cloth simulation filtering algorithm, with a preset point cloud height of Z and a simulated cloth node height of [missing value]. Solving iteratively , obtain ground point clouds ;in Indicates terrain elevation. Indicates the maximum elevation difference threshold; Ground point clouds The difference is a grid DEM, which yields the landslide digital elevation model. ,in Represents grid points; This represents the elevation value of the l-th known ground point. represents the kriging weight of the l-th known point to the current grid point, and L represents the number of known ground points used for interpolation; This indicates the digital elevation model of the landslide at grid points. The elevation value at that location.

[0025] It should be noted that the round-trip time via laser... Given the speed of light, calculate the corresponding laser ranging value R, whose transformation matrix T satisfies: ,in This indicates the number of corresponding point pairs used for registration. This represents the i2th point (source point) in the radar point cloud. Indicating the relationship between DSM point cloud and The corresponding nearest point.

[0026] According to an embodiment of the present invention, the step of acquiring three-dimensional scanning data of the well / excavation face within the landslide area and constructing a digital elevation model of the well / excavation face in conjunction with the landslide digital elevation model specifically includes: Based on the 3D scanning data of the wells / excavation faces within the landslide area, the point cloud of the wells / excavation faces was determined. ; Extract the boundary point cloud B from the DEM to determine the wellhead / excavation face, and then combine the point cloud of the wellhead / excavation face. Projecting along the well wall direction onto the profile plane yields a two-dimensional elevation model of the profile. ; Projected coordinates are Where v and w are the basis vectors of the profile plane. The origin; By combining the profile elevation model with the landslide digital elevation model, a digital elevation model of the well / excavation face is obtained. Its formula is .

[0027] It should be noted that, Let (u, v) represent the coordinates of point k on the two-dimensional profile plane; (v, w) represent a set of orthogonal basis vectors of the profile plane. This represents the original coordinates of point k in three-dimensional space. This indicates the coordinates of the selected origin point on the cross-sectional plane.

[0028] According to an embodiment of the present invention, the step of extracting the thickness of different soil types in the digital elevation model of the well / excavation face specifically includes: Extract the feature vector of each point cloud in the digital elevation model of the well / excavation face; The feature vectors of the point cloud are divided according to preset rules to obtain soil layer categories; The point cloud in the digital elevation model of the well / excavation face is divided according to different soil layer categories, and a set of depth values ​​of all points in the corresponding layer is constructed. ; Let the soil layer thickness be H. If the well / excavation face is vertical, the formula is: ,in This represents the depth value of point a in soil layer A; This indicates the thickness of soil layer A; If the well / excavation face is not vertical, then extract the dip angle of the well / excavation face. Its formula is .

[0029] It should be noted that the feature vector of the point cloud includes the color and reflection intensity of the corresponding point cloud. For example, through cluster analysis of the feature vector, the soil layer is classified into topsoil, silty clay and gravel layer. Different cohesion weighting coefficients are set between different soil layers. For example, the cohesion weighting coefficient of topsoil is set to 0.8.

[0030] According to an embodiment of the present invention, the formula for comprehensively analyzing the thickness and crack parameters of different soil types to evaluate the stability parameters of the current landslide geology is as follows: For each exploration well / excavation face, match the cracks and corresponding crack parameters within the specified area; The well / excavation face is numbered 'b', and the stability parameter of the landslide where the well / excavation face is located is set as follows: Its formula is ,in Let N represent the cohesion weighting coefficient of soil layer A, and let N represent the set of all soil layer types. This indicates the length of crack J within the defined area. This indicates the width of crack J within the defined area; Let J be the angle between the direction of crack J's extension and the slope, and M be the set of cracks within the defined area. After traversing all wells / excavation faces and numbering them, the set of stability parameters for landslides in all grids containing wells / excavation faces is obtained; The stability parameters of the landslide are obtained by averaging the values ​​of the stability parameters of the grid containing all exploration wells / excavation faces.

[0031] It should be noted that, as Figure 3 As shown, the landslide-affected area is divided into multi-grid areas, and at least one well / excavation face is set in each grid area. The direction of crack extension and the corresponding crack direction are specified. The slope direction is the main slope direction of the entire landslide. For example, the main slope direction of the landslide is 40 degrees. For example, if the angle between the direction of crack J extension and the slope direction is 5 degrees, then sin(5°) = 0.087. , This represents the corresponding weighting coefficient.

[0032] According to an embodiment of the present invention, it further includes: Compare any stability parameter in the stability parameter set with a preset first stability parameter threshold. If the stability parameter is less than the preset first stability parameter threshold, then take the grid where the corresponding stability parameter is located as the benchmark and extract the stability parameters of the landslide in the adjacent grid. The stability parameters of landslides in adjacent grids are compared and analyzed one by one with the preset second stability parameter threshold. If the stability parameters of landslides in adjacent grids are less than the preset second stability parameter threshold, a local warning message is triggered with the reference grid as the center and adjacent grids. The preset second stability parameter threshold is greater than the preset first stability parameter threshold.

[0033] It should be noted that the preset second stability parameter threshold is slightly larger than the preset first stability parameter threshold. For example, the preset first stability parameter threshold is set to 1.7 and the preset second stability parameter threshold is set to 2.0. By setting two thresholds, it is possible to prevent individual stability parameter measurement errors from misleading the entire monitoring system.

[0034] Furthermore, the preset first stability parameter threshold is dynamically adjusted based on the soil moisture content within the landslide image area. The higher the soil moisture content, the lower the corresponding preset first stability parameter threshold, thereby reducing the warning threshold.

[0035] According to an embodiment of the present invention, it further includes: Based on the landslide's impact range, determine the stability parameters of the landslide where the exploration well / excavation face is located within the grid containing the outer edge of the landslide; If the stability parameter of the landslide where the well / excavation face is located within the grid of the outer edge of the landslide is greater than the preset first stability parameter threshold and less than the preset second stability parameter threshold, then the information of the well / excavation face will be added. Based on the addition of well / excavation face information, the outer edge of the landslide is expanded outward by a preset dimension to obtain the outer edge line; Determine the outer grid area based on the outer edge line and grid lines; At least one additional well / excavation face is added within the expanded grid area to recalculate the stability parameters of the current landslide geology.

[0036] It should be noted that when the stability parameter is less than the preset first stability parameter threshold, the current landslide is in a warning state, and to ensure safety, the excavation of exploratory wells / excavation faces cannot be carried out; when the stability parameter is greater than the preset second stability parameter threshold, it means that the current landslide is very stable and no further exploration is needed; when the stability parameter is greater than the preset first stability parameter threshold and less than the preset second stability parameter threshold, it means that the current landslide is in a stable state, but not very clear. Therefore, the accuracy can be improved by increasing the number of exploratory wells / excavation faces.

[0037] According to an embodiment of the present invention, after obtaining the stability parameters of the landslide where the exploratory well / excavation face is located, the method further includes: Obtain the depth value h(t) of the exploration well / excavation face at the current time node; Obtain the depth value h(t-1) of the exploration well / excavation face at the previous time node; The depth difference is obtained by subtracting the depth of the well / excavation face at the previous time point from the depth of the well / excavation face at the current time point. Based on the preset depth range into which the depth difference falls, determine the stability parameter revision coefficient of the landslide where the corresponding exploration well / excavation face is located; Multiply the stability parameter of the landslide where the well / excavation face is located by the corresponding stability parameter revision factor to obtain the revised stability parameter of the landslide where the current well / excavation face is located.

[0038] It should be noted that by real-time monitoring of the exploratory well / excavation face and measuring the depth changes of the corresponding exploratory well / excavation face, the exploratory well / excavation face can be used as an observation point / surface to intuitively reflect whether the landslide soil is in a sliding state. For example, if the depth value at the current time point is significantly lower than the depth value at the previous time point, it indicates that the soil above the current exploratory well / excavation face is sliding down and entering the exploratory well / excavation face, thus causing the depth to decrease. Different depth differences correspond to different stability parameter revision coefficients, and the larger the depth difference, the smaller the corresponding stability parameter revision coefficient, which is less than 1.

[0039] According to an embodiment of the present invention, it further includes: Divide the depth difference by the time difference between two adjacent time points to obtain the depth backfill rate; If the depth backfill rate is greater than the preset depth backfill rate threshold, a secondary landslide warning message is generated within the grid where the corresponding well / excavation face is located. After traversing all exploration wells / excavation faces, determine the number of grids that triggered secondary landslide warning messages; If the number of grids that trigger secondary landslide warnings exceeds a preset threshold, secondary landslide warnings will be triggered for the entire monitored area.

[0040] It should be noted that the depth backfill rate directly reflects the current landslide movement status. For example, if the preset quantity threshold is 1, when the depth backfill rate of two wells / excavation faces in the grid exceeds the preset depth backfill rate threshold, a secondary landslide warning is triggered for the entire monitoring area.

[0041] This invention discloses an intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography. It acquires orthophotos and digital surface models through UAV oblique photography to identify landslide extent and surface cracks; constructs a digital elevation model of the landslide using airborne radar scanning; determines soil layer thickness by combining three-dimensional scanning of wells / excavation faces; and achieves intelligent early warning of landslide hazards by fusing multi-source data to calculate stability parameters. This invention improves the accuracy, efficiency, and automation level of landslide monitoring.

[0042] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0043] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0044] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0045] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0046] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for intelligent monitoring of landslide geological hazards based on airborne radar and oblique photography, characterized in that, include: By acquiring images of the disaster area through oblique photography by drones, constructing orthophoto images and digital surface models, and intelligently identifying the landslide impact range and surface crack distribution map based on orthophoto images; Airborne radar is used to perform three-dimensional scanning of the landslide impact area to obtain point cloud data of the landslide area, and a digital elevation model of the landslide is constructed by combining the digital surface model. Obtain 3D scanning data of wells / excavation faces within the landslide area, and construct digital elevation models of wells / excavation faces by combining them with the landslide digital elevation model; The crack parameters are determined by GNSS testing of landslide cracks and landslide profile mapping. The crack parameters include at least crack length, width, and orientation. Extract the thickness of different soil types from the digital elevation model of the well / excavation face; conduct a comprehensive analysis of the thickness and crack parameters of different soil types to assess the stability parameters of the current landslide geology; If the current stability parameters of the landslide geology are less than the preset first stability parameter threshold, an alarm message will be triggered.

2. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography according to claim 1, characterized in that, The steps for acquiring disaster area images through drone oblique photography and constructing a digital surface model specifically include: Using a drone equipped with a five-lens camera to acquire multi-view image sets Each image i contains interior and exterior orientation elements: coordinates of the camera center. and rotation matrix ; Connection points between images are established by matching preset feature points, and all parameters are optimized using bundle adjustment for any image point. Its corresponding object point They satisfy the collinearity equation: ,in Here, f is the scaling factor, and f is the focal length. This represents a 3x3 rotation matrix. The objective function for adjustment optimization is: (The vector represents the three-dimensional coordinates of the camera center.) ; Based on minimizing the energy function, a disparity map is determined, and a dense point cloud is generated from the photographic set. ; The dense point cloud is interpolated onto a predefined rule network to obtain a digital surface model.

3. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography according to claim 2, characterized in that, The steps for constructing the digital elevation model of the landslide specifically include: Airborne LiDAR systems measure laser round-trip time. Then, point cloud data is obtained by combining it with POS data. ,in ,in , This refers to the laser emission angle; This represents the three-dimensional coordinates of the GNSS antenna phase center, where R represents the laser ranging value. This represents the attitude rotation matrix provided by the inertial measurement unit. radar point cloud Point clouds in digital surface models Registration yields the transformation matrix T; Ground points are separated based on a preset cloth simulation filtering algorithm, with a preset point cloud height of Z and a simulated cloth node height of [missing value]. Solving iteratively , obtain ground point clouds ;in Indicates terrain elevation. Indicates the maximum elevation difference threshold; Ground point clouds The difference is a grid DEM, which yields the landslide digital elevation model. ,in Represents grid points; This represents the elevation value of the l-th known ground point. represents the kriging weight of the l-th known point to the current grid point, and L represents the number of known ground points used for interpolation; This indicates the digital elevation model of the landslide at grid points. The elevation value at that location.

4. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography according to claim 3, characterized in that, The steps of acquiring three-dimensional scanning data of the well / excavation face within the landslide area and constructing a digital elevation model of the well / excavation face in conjunction with the landslide digital elevation model specifically include: Based on the 3D scanning data of the wells / excavation faces within the landslide area, the point cloud of the wells / excavation faces was determined. ; Extract the boundary point cloud B from the DEM to determine the wellhead / excavation face, and then combine the point cloud of the wellhead / excavation face. Projecting along the well wall direction onto the profile plane yields a two-dimensional elevation model of the profile. ; Projected coordinates are Where v and w are the basis vectors of the profile plane. The origin; By combining the profile elevation model with the landslide digital elevation model, a digital elevation model of the well / excavation face is obtained. Its formula is .

5. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography as described in claim 4, characterized in that, The steps for extracting the thickness of different soil types in the digital elevation model of the well / excavation face specifically include: Extract the feature vector of each point cloud in the digital elevation model of the well / excavation face; The feature vectors of the point cloud are divided according to preset rules to obtain soil layer categories; The point cloud in the digital elevation model of the well / excavation face is divided according to different soil layer categories, and a set of depth values ​​of all points in the corresponding layer is constructed. ; Let the soil layer thickness be H. If the well / excavation face is vertical, the formula is: ,in This represents the depth value of point a in soil layer A; This indicates the thickness of soil layer A; If the well / excavation face is not vertical, then extract the dip angle of the well / excavation face. Its formula is .

6. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography as described in claim 5, characterized in that, The formula for comprehensively analyzing the thickness and crack parameters of different soil types to evaluate the stability parameters of the current landslide geology is as follows: For each exploration well / excavation face, match the cracks and corresponding crack parameters within the specified area; The well / excavation face is numbered 'b', and the stability parameter of the landslide where the well / excavation face is located is set as follows: Its formula is ,in Let N represent the cohesion weighting coefficient of soil layer A, and let N represent the set of all soil layer types. This indicates the length of crack J within the defined area. This indicates the width of crack J within the defined area; Let J be the angle between the direction of crack J's extension and the slope, and M be the set of cracks within the defined area. After traversing all wells / excavation faces and numbering them, the set of stability parameters for landslides in all grids containing wells / excavation faces is obtained; The stability parameters of the landslide are obtained by averaging the values ​​of the stability parameters of the grid containing all exploration wells / excavation faces.

7. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography according to claim 6, characterized in that, Also includes: Compare any stability parameter in the stability parameter set with a preset first stability parameter threshold. If the stability parameter is less than the preset first stability parameter threshold, then take the grid where the corresponding stability parameter is located as the benchmark and extract the stability parameters of the landslide in the adjacent grid. The stability parameters of landslides in adjacent grids are compared and analyzed one by one with the preset second stability parameter threshold. If the stability parameters of landslides in adjacent grids are less than the preset second stability parameter threshold, a local warning message is triggered with the reference grid as the center and adjacent grids. The preset second stability parameter threshold is greater than the preset first stability parameter threshold.

8. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography according to claim 7, characterized in that, Also includes: Based on the landslide's impact range, determine the stability parameters of the landslide where the exploration well / excavation face is located within the grid containing the outer edge of the landslide; If the stability parameter of the landslide where the well / excavation face is located within the grid of the outer edge of the landslide is greater than the preset first stability parameter threshold and less than the preset second stability parameter threshold, then the information of the well / excavation face will be added. Based on the addition of well / excavation face information, the outer edge of the landslide is expanded outward by a preset dimension to obtain the outer edge line; Determine the outer grid area based on the outer edge line and grid lines; At least one additional well / excavation face is added within the expanded grid area to recalculate the stability parameters of the current landslide geology.

9. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography according to claim 6, characterized in that, After obtaining the stability parameters of the landslide where the exploratory well / excavation face is located, the following are also included: Obtain the depth value h(t) of the exploration well / excavation face at the current time node; Obtain the depth value h(t-1) of the exploration well / excavation face at the previous time node; The depth difference is obtained by subtracting the depth of the well / excavation face at the previous time point from the depth of the well / excavation face at the current time point. Based on the preset depth range into which the depth difference falls, determine the stability parameter revision coefficient of the landslide where the corresponding exploration well / excavation face is located; Multiply the stability parameter of the landslide where the well / excavation face is located by the corresponding stability parameter revision factor to obtain the revised stability parameter of the landslide where the current well / excavation face is located.

10. The intelligent monitoring method for landslide geological hazards based on airborne radar and oblique photography according to claim 9, characterized in that, Also includes: Divide the depth difference by the time difference between two adjacent time points to obtain the depth backfill rate; If the depth backfill rate is greater than the preset depth backfill rate threshold, a secondary landslide warning message is generated within the grid where the corresponding well / excavation face is located. After traversing all exploration wells / excavation faces, determine the number of grids that triggered secondary landslide warning messages; If the number of grids that trigger secondary landslide warnings exceeds a preset threshold, secondary landslide warnings will be triggered for the entire monitored area.

Images