Dam slope monitoring method and system based on radar interference and aerial survey data

Abstract

The invention provides a dam slope monitoring method and system based on radar interference and aerial survey data, and relates to the technical field of deformation measurement. The method comprises the following steps: acquiring radar interference measurement data of a target dam slope area, and carrying out multi-temporal registration and phase difference processing to obtain displacement time sequence data; determining displacement change data based on the displacement time sequence data, and determining a first risk area according to the displacement change data; controlling the flight platform to collect aerial survey data of the first risk area; performing feature extraction on the image data to obtain an image structure feature map, and performing terrain gridding modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map; performing coordinate alignment on the image structure feature map and the terrain curvature distribution map to obtain a combined feature map; and determining a risk scoring value by using the risk scoring function, and determining a second risk area based on the risk scoring value. According to the scheme, the accuracy of dam slope anomaly recognition can be improved while the monitoring coverage is ensured.

Description

Background Art

[0002] At present, for the high slope areas of dams in areas such as reservoirs, flood discharge areas, and along water diversion projects, usually two methods are adopted for slope monitoring. One is the contact monitoring method of arranging monitoring devices at fixed monitoring points. For example, total stations, global navigation satellite system receivers and other devices are used to conduct periodic observations on key positions of the slope. Although this method has high accuracy, the number of points is limited and the layout is not flexible, making it difficult to achieve large-scale coverage. Another method is the non-contact monitoring technology based on synthetic aperture radar interferometry (InSAR), which can obtain surface deformation information within a large area range. However, the method based on InSAR has certain limitations in the resolution ability of surface detail changes under complex terrain conditions, and it is difficult to achieve high-precision identification of small-scale local anomalies.

[0003] Therefore, how to improve the accuracy of anomaly identification while ensuring monitoring coverage has become an urgent problem to be solved in the current dam slope monitoring technology. Summary of the Invention

[0004] The purpose of the embodiments of the present disclosure is to provide a dam slope monitoring method based on radar interference and aerial survey data, a dam slope monitoring system based on radar interference and aerial survey data, an electronic device, and a computer-readable storage medium, which can combine temporal displacement information with topographic structure characteristics to achieve rapid identification and precise positioning of the abnormal risks of the dam slope while ensuring monitoring coverage, and can also improve the accuracy of anomaly detection.

[0005] Other features and advantages of the present disclosure will become apparent through the following detailed description, or be learned in part through the practice of the present disclosure.

[0006] According to the first aspect of the embodiments of the present disclosure, a dam slope monitoring method based on radar interferometry and aerial survey data is provided, including: acquiring radar interferometry data of a target dam slope area at multiple time points, performing multi-temporal registration and phase difference processing on the radar interferometry data to obtain displacement time series data; determining displacement change data of each measurement position in a preset observation period based on the displacement time series data, and determining a first risk area according to the displacement change data; determining an aerial survey path according to the first risk area, and controlling a flight platform to collect aerial survey data of the first risk area based on the aerial survey path, where the aerial survey data includes image data and point cloud data; extracting features from the image data to obtain an image structure feature map, performing terrain grid modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map; aligning the coordinates of the image structure feature map and the terrain curvature distribution map to obtain a combined feature map; using a risk scoring function to determine risk score values corresponding to each grid in the combined feature map, and determining a second risk area based on the risk score values.

[0007] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the performing multi-temporal registration and phase difference processing on the radar interferometry data to obtain displacement time series data includes: performing image registration processing on the radar interferometry data based on orbital parameters and georeference coordinates to obtain a multi-temporal radar image set; performing coherence screening and phase difference calculation on the multi-temporal radar image set to obtain phase time series data corresponding to multiple phase stable points; constructing the displacement time series data for describing surface deformation based on the position coordinates of the phase stable points and the phase time series data.

[0008] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the performing coherence screening and phase difference calculation on the multi-temporal radar image set to obtain phase time series data corresponding to multiple phase stable points includes: determining the interference phase difference between each image pixel in the multi-temporal radar image set at adjacent time points, and determining the coherence index value of the corresponding image pixel based on the interference phase difference; comparing the coherence index value with a preset coherence threshold, and screening out multiple phase stable points from the image pixels according to the comparison result; constructing the phase time series data based on the interference phase difference corresponding to the phase stable points.

[0009] In some exemplary embodiments of the present disclosure, based on the foregoing solution, determining displacement change data of each measurement position within a preset observation period based on the displacement time series data, and determining a first risk area according to the displacement change data includes: determining the displacement amount and displacement rate of each measurement position in the displacement time series data within the preset observation period; screening out first risk positions from the measurement positions based on the displacement amount and displacement rate; and generating the first risk area based on the position distribution of the first risk positions.

[0010] In some exemplary embodiments of the present disclosure, based on the foregoing solution, determining an aerial survey path according to the first risk area includes: converting the image pixel coordinates corresponding to the measurement positions in the first risk area into geospatial coordinates through coordinate calculation and spatial projection processing; and determining at least one aerial survey path based on the geospatial coordinates.

[0011] In some exemplary embodiments of the present disclosure, based on the foregoing solution, extracting features from the image data to obtain an image structure feature map includes: performing gray conversion on the image data to obtain an initial gray image; using the three-dimensional terrain model corresponding to the image data to perform gray enhancement processing on the initial gray image to obtain optimized image data; performing edge feature extraction and gray change feature extraction on the optimized image data, and generating the image structure feature map based on the feature extraction results.

[0012] In some exemplary embodiments of the present disclosure, based on the foregoing solution, using the three-dimensional terrain model corresponding to the image data to perform gray enhancement processing on the initial gray image to obtain optimized image data includes: constructing a three-dimensional terrain model based on the point cloud data of the area corresponding to the image data, and determining the slope normal direction corresponding to each pixel point in the initial gray image based on the three-dimensional terrain model; determining the light incident direction vector corresponding to the image data according to the acquisition time and geographical location of the image data; calculating angles based on the slope normal direction and the light incident direction vector to generate the light incident angle value corresponding to each pixel point; and performing gray enhancement on the initial gray image based on the light incident angle value to obtain the optimized image data.

[0013] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the gray-scale enhancement of the initial gray-scale image based on the light incident angle value to obtain the optimized image data includes: dividing the light incident angle values corresponding to the pixel points in the initial gray-scale image into corresponding angle intervals, where the angle intervals include a strong light interval, a medium light interval, a backlight interval, and a shadow interval; using the gray-scale remapping function corresponding to each angle interval to perform partitioned gray-scale enhancement processing on the initial gray-scale image to obtain the optimized image data; wherein, the gray-scale remapping function includes any one of a linear mapping function, a gamma correction function, and a logarithmic enhancement function.

[0014] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the gray-scale enhancement of the initial gray-scale image based on the light incident angle value to obtain the optimized image data includes: determining the enhancement weight corresponding to each pixel point based on the light incident angle value corresponding to the pixel points in the initial gray-scale image, where the enhancement weight is proportional to the cosine value of the light incident angle value; performing weighted enhancement processing on the original gray-scale values of the pixel points in the initial gray-scale image based on the enhancement weight to generate the optimized image data.

[0015] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the terrain grid modeling and curvature calculation of the point cloud data to obtain a terrain curvature distribution map includes: constructing a local neighborhood corresponding to each data point in the point cloud data and determining the point cloud density and normal variation value corresponding to the local neighborhood; determining the grid size corresponding to each data point based on the point cloud density and normal variation value; performing non-uniform grid division on the point cloud data according to the grid size to generate a three-dimensional terrain model; calculating the curvature of each grid node in the three-dimensional terrain model and generating the terrain curvature distribution map according to the curvature calculation result.

[0016] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the use of the risk scoring function to determine the risk scoring value corresponding to each grid in the joint feature map includes: respectively normalizing the structural feature data and terrain curvature in the joint feature map to obtain structure-normalized data and curvature-normalized data; using the risk scoring function constructed by the weighting coefficients to perform weighted calculation on the structure-normalized data and the curvature-normalized data to obtain the risk scoring value corresponding to each grid.

[0017] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the above-mentioned dam slope monitoring method based on radar interferometry and aerial survey data further includes: in response to the second risk area being a surface anomaly area, controlling a survey robot to obtain actual deformation data and environmental data of at least one target location in the second risk area; wherein, the actual deformation data includes any one or more of crack width, crack length, and step height, and the environmental data includes any one or more of air humidity, shallow water content, and temperature data.

[0018] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the above-mentioned dam slope monitoring method based on radar interferometry and aerial survey data further includes: in response to the second risk area being a deformation area, obtaining historical displacement time series data corresponding to the deformation area; performing multi-time scale deformation prediction based on the historical displacement time series data to obtain a deformation prediction result of the deformation area.

[0019] In some exemplary embodiments of the present disclosure, based on the foregoing solution, performing multi-time scale deformation prediction based on the historical displacement time series data to obtain a deformation prediction result of the deformation area includes: using a first time scale and a second time scale to respectively segment the historical displacement time series data to obtain a first-scale subsequence and a second-scale subsequence, the first time scale being less than the second time scale; using the first-scale subsequence to perform short-term deformation prediction to obtain a first prediction result; using the second-scale subsequence to perform long-term deformation prediction to obtain a second prediction result; and performing weighted fusion on the first prediction result and the second prediction result to obtain a deformation prediction result of the deformation area.

[0020] According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including: a processor; and a memory, on which computer-readable instructions are stored, and when the computer-readable instructions are executed by the processor, the dam slope monitoring method based on radar interferometry and aerial survey data as in the first aspect is implemented.

[0021] According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the dam slope monitoring method based on radar interferometry and aerial survey data as in the first aspect is implemented.

[0022] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects: In the dam slope monitoring method based on radar interferometry and aerial survey data in the exemplary embodiments of the present disclosure, on the one hand, by acquiring radar interferometry data of the target dam slope area at multiple time points and performing multi-temporal registration and phase difference processing on it, the deformation changes on the slope surface can be accurately measured. On the other hand, based on the displacement time series data, the displacement change data of each measurement position in the preset observation period is determined, and the first risk area is identified thereby, effectively avoiding the limitations of inflexible monitoring layout at fixed points and difficulty in achieving wide coverage, and improving the ability to early identify abnormal changes in the slope measurement area. On the further hand, by planning the aerial survey path according to the first risk area and collecting image data and point cloud data, the fine inspection of the key risk area by the flight platform is realized. Combining image feature extraction and point cloud curvature analysis, the image structure feature map and the terrain curvature distribution map are further obtained, so as to characterize the terrain features in two dimensions of structural information and terrain curvature. Further, the risk score function is used to determine the risk score value corresponding to each grid in the joint feature map, and the second risk area is determined based on the risk score value, which can accurately identify local abnormal areas by integrating image structure and terrain curvature features. That is to say, the embodiments of the present disclosure can accurately identify and precisely locate the abnormal risks of the dam slope while ensuring monitoring coverage.

[0023] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0025] Figure 1 Schematically shows a flowchart of a dam slope monitoring method based on radar interferometry and aerial survey data according to some embodiments of the present disclosure.

[0026] Figure 2 Schematically shows a flowchart of obtaining phase time series data according to some embodiments of the present disclosure.

[0027] Figure 3 Schematically shows a flowchart of generating the first risk area according to some embodiments of the present disclosure.

[0028] Figure 4 Schematically shows a flowchart of obtaining optimized image data according to some embodiments of the present disclosure.

[0029] Figure 5 Schematically shows a block diagram of a dam slope monitoring system based on radar interferometry and aerial survey data according to some embodiments of the present disclosure.

[0030] Figure 6 Schematically shows a structural diagram of a computer system of an electronic device according to some embodiments of the present disclosure.

[0031] Figure 7 Schematically shows a diagram of a computer-readable storage medium according to some embodiments of the present disclosure.

[0032] In the drawings, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Description of Specific Embodiments

[0033] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. On the contrary, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.

[0034] The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this specification. The singular forms "a", "the", and "said" used in this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items. Now, the exemplary embodiments will be described more fully with reference to the drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and comprehensive, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art.

[0035] In addition, the drawings are only schematic diagrams and are not necessarily drawn to scale. The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0036] In the present exemplary embodiment, first, a dam slope monitoring method based on radar interferometry and aerial survey data is provided. Figure 1Schematically shows a flowchart of a dam slope monitoring method based on radar interferometry and aerial survey data according to some embodiments of the present disclosure. Refer to Figure 1 As shown, the dam slope monitoring method based on radar interferometry and aerial survey data may include the following steps: Step S110: Obtain radar interferometry measurement data of the target dam slope area at multiple time points, perform multi-temporal registration and phase difference processing on the radar interferometry measurement data to obtain displacement time series data; Step S120: Determine the displacement change data of each measurement position during a preset observation period based on the displacement time series data, and determine the first risk area according to the displacement change data; Step S130: Determine the aerial survey path according to the first risk area, and control the flight platform to collect aerial survey data of the first risk area based on the aerial survey path. The aerial survey data includes image data and point cloud data; Step S140: Extract features from the image data to obtain an image structure feature map, perform terrain grid modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map; Step S150: Align the coordinates of the image structure feature map and the terrain curvature distribution map to obtain a combined feature map; Step S160: Use a risk scoring function to determine the risk score value corresponding to each grid in the combined feature map, and determine the second risk area based on the risk score value.

[0037] According to the dam slope monitoring method based on radar interferometry and aerial survey data in this exemplary embodiment, on the one hand, by obtaining radar interferometry measurement data of the target dam slope area at multiple time points and performing multi-temporal registration and phase difference processing on it, the deformation changes on the slope surface can be accurately measured. On the other hand, based on the displacement time series data, the displacement change data of each measurement position during the preset observation period is determined, and the first risk area is identified thereby, effectively avoiding the limitations of inflexible layout of fixed-point monitoring and difficulty in achieving large-scale coverage, and improving the early identification ability of abnormal changes in the slope measurement area. On the other hand, by planning the aerial survey path according to the first risk area and collecting image data and point cloud data, fine inspection of key risk areas is realized using the flight platform. Combining image feature extraction and point cloud curvature analysis, an image structure feature map and a terrain curvature distribution map are further obtained, thereby characterizing the terrain features in two dimensions of structural information and terrain curvature. Further, using a risk scoring function to determine the risk score value corresponding to each grid in the combined feature map and determining the second risk area based on the risk score value can accurately identify local abnormal areas by integrating image structure and terrain curvature features. Furthermore, this embodiment can accurately identify and precisely locate the abnormal risks of the dam slope while ensuring monitoring coverage.

[0038] Next, the dam slope monitoring method based on radar interferometry and aerial survey data in this exemplary embodiment will be further described.

[0039] Step S110: Obtain radar interferometry data of the target dam slope area at multiple time points, perform multi-temporal registration and phase difference processing on the radar interferometry data to obtain displacement time series data.

[0040] Among them, the dam slope area can represent the natural and artificial slope areas on the side or downstream side of the hydraulic structure, which often have risks such as deformation and cracks. The hydraulic structure can be a reservoir, a flood discharge area, and along the water diversion project, etc. The radar interferometry data can represent the radar image data with phase information obtained by a measurement platform with an interferometric synthetic aperture radar (InSAR) function at multiple observation times. Multi-temporal registration can represent the operation of uniformly aligning the radar interferometry data obtained at multiple observation times in the spatial reference system. This registration process can achieve spatial consistency between images based on orbital parameters, geographical reference coordinates, or feature point matching. Phase difference processing can represent the processing operation of performing pixel-by-pixel difference calculation on the interferometric phase data in the registered multi-temporal radar images, which is used to calculate the relative phase change between different observation time points. This phase change can reflect the displacement change of the measurement point in the radar line-of-sight direction. The displacement time series data can represent a set of displacement measurement data arranged in the order of observation time.

[0041] Step S120: Determine the displacement change data of each measurement position in the preset observation period based on the displacement time series data, and determine the first risk area according to the displacement change data.

[0042] Among them, the displacement change data can represent a set of measurement data calculated based on the displacement time series data and used to describe the relative displacement change amount of each measurement position in the monitoring area between consecutive observation periods. The first risk area can represent the abnormal area identified according to the displacement change data and showing abnormal displacement increment or displacement rate exceeding the limit. The displacement change data can reflect the terrain change of the target dam slope area. When the slope deforms or cracks appear, it can be reflected through the displacement change data. However, due to the insufficient accuracy of the radar interferometry data, only the preliminary screening of the abnormal area in the dam slope surface can be achieved by obtaining the first risk area. Therefore, subsequent aerial survey data collection is required for secondary identification.

[0043] Step S130: Determine the aerial survey path according to the first risk area, and control the flight platform to collect the aerial survey data of the first risk area based on the aerial survey path. The aerial survey data includes image data and point cloud data.

[0044] Among them, the aerial survey path can represent a three-dimensional trajectory line for guiding a flight platform to perform a measurement task in the first risk area. The flight platform can represent an aerial survey device equipped with a sensor device and capable of autonomously flying according to a preset flight path, and its forms can include unmanned aerial vehicles, fixed-wing aircraft, rotary-wing aircraft, and other mobile devices with the ability to collect data in the air. The image data can represent visible light image data collected by the flight platform and covering the first risk area. The point cloud data can be a set of measurement data composed of three-dimensional space coordinate points.

[0045] Step S140: Extract features from the image data to obtain an image structure feature map, and perform terrain grid modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map.

[0046] Among them, the image structure feature map can represent a feature map generated based on the image data through a feature extraction algorithm and used to characterize structural changes such as the texture, edges, and cracks on the slope surface. Terrain grid modeling can represent performing spatial division and topological connection processing on the point cloud data to construct a three-dimensional grid model that can characterize the terrain of the dam slope. The terrain curvature distribution map can represent a spatial distribution map formed by calculating geometric curvature values at each grid node or surface patch based on the terrain grid modeling results. By obtaining the image structure feature map and the terrain curvature distribution map, it is possible to provide multi-source heterogeneous data for the accurate identification of subsequent local abnormal areas, thereby improving the accuracy of identification.

[0047] Step S150: Align the coordinates of the image structure feature map and the terrain curvature distribution map to obtain a combined feature map.

[0048] Coordinate alignment can represent mapping the image structure feature map and the terrain curvature distribution map to a unified reference coordinate system to achieve the spatial consistency of the corresponding position data. The combined feature map can represent a composite feature map constructed by splicing the structural information in the image structure feature map and the geometric information in the terrain curvature distribution map according to the spatial position.

[0049] In this embodiment, the combined feature map can be generated through the following steps: Extract the spatial reference coordinate information corresponding to each spatial grid in the image structure feature map and the terrain curvature distribution map; Based on the spatial reference coordinate information, perform grid pairing operations on the image structure feature map and the terrain curvature distribution map in a unified geospatial coordinate system to establish a spatial mapping relationship between the image structure features and the terrain curvature features; For each paired spatial position, splice the structural feature data in the image structure feature map and the curvature data in the terrain curvature distribution map position by position to generate a combined feature map. Among them, the combined feature map is used to jointly represent structural information and terrain information in the same spatial grid.

[0050] Step S160: Use a risk scoring function to determine the risk score values corresponding to each grid in the joint feature map, and determine the second risk area based on the risk score values.

[0051] Among them, the risk scoring function can represent a function used to comprehensively evaluate the structural features and terrain features contained in each spatial grid position in the joint feature map. The risk score value can represent the quantitative scoring result calculated by the risk scoring function for each spatial grid in the joint feature map. The second risk area can represent an abnormal area with slope terrain risks such as cracks and deformations further identified based on the risk score values within the spatial range corresponding to the joint feature map, that is, the first risk area. By obtaining the second risk area, it is possible to achieve secondary accurate identification of the risks of cracks and deformations within the first risk area, improving the positioning accuracy and discrimination reliability of slope anomaly measurement.

[0052] Next, the content in the above steps S110 to S160 will be described in detail.

[0053] In some embodiments, multi-temporal registration and phase difference processing are performed on the radar interferometry data to obtain displacement time series data, which specifically includes the following technical steps: Based on the orbital parameters and georeference coordinates, perform image registration processing on the radar interferometry data to obtain a multi-temporal radar image set; perform coherence screening and phase difference calculation on the multi-temporal radar image set to obtain the phase time series data corresponding to multiple phase stable points; based on the position coordinates of the phase stable points and the phase time series data, construct displacement time series data for describing surface deformation.

[0054] Among them, the orbital parameters can represent a set of information used to characterize the spatial position and attitude state of the remote sensing platform carrying the synthetic aperture radar in orbit, and the orbital parameters include but are not limited to orbital altitude, trajectory, speed, attitude angle, and observation time, etc. The georeference coordinates can represent a coordinate system used to map the pixel points in the radar image to the actual geographical space, such as the geodetic coordinate system or the projection coordinate system, etc. The multi-temporal radar image set can represent a set of radar images obtained at different observation times and having a unified geographical space reference after registration processing. In specific implementation, the orbital parameter and georeference coordinate information contained in the radar interferometry data can be extracted first, and a spatial reference model corresponding to the target dam slope area can be constructed. Then, based on the spatial reference model, coordinate transformation processing is performed on the radar images obtained at different times to correct the image geometric distortion caused by sensor viewing angle, attitude change, or earth curvature. Finally, under the unified geographical space coordinate system, spatial registration processing is performed on the pixel points in the radar image to eliminate the spatial offset and image misalignment between multi-temporal data, and a multi-temporal radar image set is obtained.

[0055] Furthermore, coherence screening can represent a processing procedure for evaluating the phase consistency of each pixel point in multi-temporal radar images. By calculating the interference coherence index between images at adjacent times, pixel points with high coherence and stable signal-to-noise ratio are screened out. Phase difference calculation can represent the differential operation of the phase information of radar images obtained at different time points at the same spatial position to obtain the relative phase change data in the time dimension. Phase stable points can represent a set of pixel points in multi-temporal radar images that have long-term stable coherence and continuous phase changes. Phase time series data can represent the sequence of interference phase change values arranged in the order of observation time for specific phase stable points.

[0056] When constructing displacement time series data based on the position coordinates of phase stable points and phase time series data, the following steps can be taken: Extract the position coordinates of multiple phase stable points in a unified geographic coordinate system; perform phase unwrapping on the phase time series data corresponding to the phase stable points and convert it into a continuous phase change sequence with physical meaning; calculate the surface displacement values of the phase stable points at multiple observation times based on the continuous phase change sequence and the radar wavelength parameter; combine the surface displacement values in chronological order to construct the displacement time series data for the corresponding measurement points. Exemplarily, the displacement time series data can be expressed as: where represents the displacement time series corresponding to the th phase stable point, represents the surface displacement value of this phase stable point at the th observation time , represents the th moment, satisfying , constituting an equally spaced or unequally spaced observation time series.

[0057] In this embodiment, by screening phase stable points with long-term coherence and constructing displacement time series data based on their positions and phase time series, high-precision and low-noise measurement of surface deformation in the dam slope area is achieved, significantly improving the stability and measurement credibility of time-series deformation information in complex environments.

[0058] In some embodiments, as shown in Figure 2 , the coherence screening and phase difference calculation of the multi-temporal radar image set can be performed through steps S210 to S230 to obtain the phase time series data corresponding to multiple phase stable points, specifically including: Step S210, determine the interference phase difference between each image pixel in the multi-temporal radar image set at adjacent time points, and determine the coherence index value of the corresponding image pixel based on the interference phase difference.

[0059] The interference phase difference can represent the difference in phase values ​​generated by the echo signals of the corresponding pixel points between radar images acquired at different observation times for the same image pixel position, and the phase difference can reflect the relative displacement information of the corresponding position between the two observation time points. The coherence index value can represent a numerical measure for measuring the stability of the phase change of image pixels in multi-temporal interference images.

[0060] Exemplarily, the interference phase difference and the coherence index value may be obtained by the following steps: First, for the radar images corresponding to two adjacent time points in the multi-temporal radar image set, the image pixel positions are extracted The complex value pair , calculate the interference phase difference between adjacent time points at this position , whose expression is: in, , Respectively represent With The complex pixel value at the time, express The complex conjugate of Represents the phase angle extraction operation of a complex number.

[0061] Then, for the image pixel position Interference phase difference in all time periods Perform statistical analysis and calculate the coherence index value of the position , the coherence index value is determined by the following formula: in, represents the number of radar images, Used to measure the stability of adjacent phase differences.

[0062] Step S220 , comparing the coherence index value with a preset coherence threshold, and selecting a plurality of phase stable points from the image pixels according to the comparison result.

[0063] The coherence threshold may represent a numerical limit for determining whether an image pixel satisfies the interferometric phase stability. For example, the coherence threshold may be set to a real value between 0.3 and 0.6. When the coherence threshold is set to 0.3, more pixels may be covered, which is beneficial to increase the point density of the monitoring area. When the coherence threshold is set to 0.5 or above, the selected points have higher phase consistency.

[0064] Step S230: Construct phase-time series data based on the interference phase difference corresponding to the phase stable points.

[0065] In some embodiments, refer to Figure 3 As shown, based on the displacement-time series data, determine the displacement change data of each measurement location during the preset observation period, and determine the first risk area according to the displacement change data. The specific technical steps are as follows: Step S310: Determine the displacement amount and displacement rate of each measurement location in the displacement-time series data during the preset observation period.

[0066] Among them, the observation period can represent a preset continuous time period for observing the topographic anomalies in the dam slope area. Exemplarily, the values of the observation period can be 6 days, 12 days, 24 days, 30 days, and other appropriate values. It can also be set according to the repeat orbit period of the used interferometric radar satellite system. In specific implementation, multiple surface displacement values of each measurement location during the preset observation period can be extracted, and then statistical processing is performed on the multiple surface displacement values to calculate the displacement amount of each measurement location during each preset observation period. Finally, based on the cumulative displacement amount and the total time length of the preset observation period, calculate the displacement rate of the measurement location during this observation period.

[0067] Step S320: Based on the displacement amount and displacement rate, screen out the first risk locations from the measurement locations.

[0068] Specifically, the displacement amount of each measurement location during the preset observation period can be compared with the displacement threshold, and the measurement locations with displacement amounts exceeding the displacement threshold are marked. The displacement rate of each measurement location during the preset observation period is compared with the rate threshold, and the measurement locations with displacement rates exceeding the second threshold are marked. According to the marking results, form the first risk location set.

[0069] Step S330: Generate the first risk area based on the location distribution of the first risk locations.

[0070] Exemplarily, when generating the first risk area based on the location distribution of the first risk locations, the following steps can be taken: Extract the spatial location coordinates of the first risk locations in the geographic coordinate system; perform adjacency analysis on the spatial location coordinates to identify multiple groups of risk locations that are connected to each other or have a distance less than the preset distance threshold; based on each group of risk locations, construct multiple closed polygon areas covering their boundaries; finally, use all the closed polygon areas as the first risk area.

[0071] In some embodiments, the aerial survey path is determined according to the first risk area, and the specific technical steps are as follows: Through coordinate calculation and spatial projection processing, the image pixel coordinates corresponding to the measurement positions in the first risk area are converted into geospatial coordinates; At least one aerial survey path is determined based on the geospatial coordinates.

[0072] Among them, coordinate calculation can represent performing position inverse calculation processing on the image pixel positions based on the exterior orientation elements of the image acquisition device and the image geometric model, and obtaining the initial projection result of the corresponding measurement position in the three-dimensional space coordinate system. Spatial projection processing can represent performing transformation processing on the initial spatial coordinate values obtained by coordinate calculation according to the preset projection coordinate system. Geospatial coordinates can represent the three-dimensional coordinates used to represent the spatial position of a specific ground measurement position under a unified geographic reference framework.

[0073] In this embodiment, when performing coordinate transformation, for the image pixel coordinates corresponding to each measurement position in the first risk area, coordinate calculation processing can be performed based on the internal parameters of the sensor at the image acquisition moment and the attitude solution result of the flight platform to obtain the corresponding three-dimensional image space position. Then, the three-dimensional image space position is input into the set geographic projection model, and spatial projection processing is performed according to the selected geographic reference coordinate system to be converted into geospatial coordinates including longitude, latitude, and elevation. The converted geospatial coordinates are used to characterize the actual spatial distribution of each measurement position in the first risk area.

[0074] In addition, since the coverage area of the dam slope area is usually large and the energy consumption of the flight platform is limited, multiple aerial survey paths can be generated to ensure the inspection efficiency. In specific implementation, based on the geospatial coordinates, the first risk area can be spatially divided, and the geospatial coordinates are divided into several independent sub-areas according to the preset coverage radius and flight range parameters. For the set of geospatial coordinates in each sub-area, a path node sequence is generated based on the spatial distance relationship between nodes, and a sub-aerial survey path matching the performance parameters of the flight platform is constructed. The multiple sub-aerial survey paths are combined in spatial order to generate at least one continuous or segmented aerial survey path covering the first risk area, which is used to control the flight platform to complete the aerial survey tasks in sequence.

[0075] In some embodiments, feature extraction is performed on the image data to obtain an image structure feature map, and the specific process is as follows: The image data is subjected to grayscale conversion to obtain an initial grayscale image; Using the corresponding three-dimensional terrain model of the image data, the initial grayscale image is subjected to grayscale enhancement processing to obtain optimized image data; Edge feature extraction and grayscale change feature extraction are performed on the optimized image data, and an image structure feature map is generated based on the feature extraction results.

[0076] Among them, gray-scale conversion can represent converting the color values of each pixel point in the color image data collected by the flying platform into corresponding single-channel gray-scale values. The three-dimensional terrain model can represent the spatial geometric model of the dam slope area constructed based on the point cloud data. Gray-scale enhancement can represent, according to the spatial orientation of each pixel in the image in the three-dimensional terrain model and in combination with the set light incident direction, performing weighted adjustment processing on the gray-scale values of the corresponding pixels in the initial gray-scale image to enhance the edge weakening areas caused by insufficient light or shadow occlusion. In this embodiment, the problem of gray-scale attenuation caused by differences in slope orientation or local shadows is effectively corrected through gray-scale enhancement, and the significance of edge structures and brightness changes in the image is enhanced. In addition, based on the optimized image data, joint extraction of edge features and gray-scale change features is performed, improving the recognition ability of crack edges and deformation areas.

[0077] In some embodiments, as shown in Figure 4 utilize the three-dimensional terrain model corresponding to the image data to perform gray-scale enhancement processing on the initial gray-scale image to obtain optimized image data, specifically including the following technical steps: Step S410, construct a three-dimensional terrain model based on the point cloud data of the corresponding area of the image data, and determine the slope normal direction corresponding to each pixel point in the initial gray-scale image based on the three-dimensional terrain model.

[0078] Among them, the slope normal direction can represent the direction information of the normal vector of the spatial surface tangent plane of each pixel point in the constructed three-dimensional terrain model in the corresponding area of the image data, and is used to characterize the terrain orientation. Exemplarily, the slope normal direction can be obtained through the following process: First, perform neighborhood search processing on the point cloud corresponding to the pixel positions of the initial gray-scale image in the three-dimensional terrain model to obtain the local point set corresponding to the pixel position, where represents the th three-dimensional point coordinate in the local point cloud, represents the number of points in the neighborhood, represents the pixel position of the point projected onto the two-dimensional image, represents the elevation value of the point. represents the elevation value of the point.

[0079] Then, based on the least squares method, perform local plane fitting on the point set to construct a fitting plane equation: Among them, , , are plane fitting parameters, represents the slope component of the fitting plane in the direction, represents the slope component of the fitting plane in the direction, represents the elevation intercept term of the fitting plane in three-dimensional space, which is calculated by minimizing the sum of squared residuals with as samples during the fitting process.

[0080] Finally, based on the above fitting plane equation , determine the slope normal direction vector corresponding to the pixel point , and its calculation formula is: where, represents the unit normal vector, and the components , , respectively represent the spatial orientation components of the slope where the pixel point is located in the direction, direction, and direction, represents the modulus of the normal vector.

[0081] Step S420, determine the light incident direction vector corresponding to the image data according to the acquisition time and geographical location of the image data.

[0082] Among them, the light incident direction vector can represent the incident direction of the solar radiation line in three-dimensional space under the conditions of a specific image acquisition time and geographical location, and is used to characterize the direction information of the sunlight irradiating the ground at the image acquisition moment.

[0083] In specific implementation, first obtain the acquisition time information and the central geographical coordinate information of the image data. Among them, the acquisition time information includes the standard timestamp corresponding to the image shooting, and the central geographical coordinate information includes the longitude and latitude coordinate values of the corresponding image area. Then, based on the acquisition time information and the central geographical coordinate information, calculate the solar altitude angle and the solar azimuth angle corresponding to the image acquisition moment. Among them, the solar altitude angle represents the angle between the central ray of the sun and the ground plane, and the solar azimuth angle represents the clockwise angle between the solar projection in the horizontal plane and the due north direction. Finally, based on the solar altitude angle and the solar azimuth angle, construct the three-dimensional light incident direction vector , where, ; ; ; represents the solar altitude angle, represents the solar azimuth angle, , , respectively represent the direction components of the incident light in the geographical eastward, northward, and vertically upward directions; Step S430: Calculate angles based on the slope normal direction and the light incident direction vector to generate the light incident angle values corresponding to each pixel. Here, the light incident angle value can be expressed as the angle between the slope normal vector and the light incident direction vector.

[0084] Step S440: Perform gray-scale enhancement on the initial gray-scale image based on the light incident angle values to obtain optimized image data. Here, by performing gray-scale enhancement on the initial gray-scale image based on the light incident angle values, the image pixels in the low incident angle region can be compensated with targeted brightness, while the pixels in the high incident angle region maintain the original brightness or are moderately adjusted, enhancing the identifiability of structural details such as cracks, faults, and deformation boundaries in the image.

[0085] In some embodiments, performing gray-scale enhancement on the initial gray-scale image based on the light incident angle values to obtain optimized image data specifically includes the following technical steps: Divide the light incident angle values corresponding to each pixel in the initial gray-scale image into corresponding angle intervals, where the angle intervals include a strong light interval, a medium light interval, a backlight interval, and a shadow interval; Use the gray-scale remapping function corresponding to each angle interval to perform partitioned gray-scale enhancement on the initial gray-scale image to obtain optimized image data; Among them, the gray-scale remapping function includes any one of a linear mapping function, a gamma correction function, and a logarithmic enhancement function.

[0086] Here, the gray-scale remapping function can be expressed as a function for performing numerical transformation on the pixel gray-scale values in the gray-scale image. Exemplarily, when using a linear mapping function as the gray-scale remapping function, the gray-scale enhancement of the initial gray-scale image can be performed through the following process: First, divide the light incident angle value θ into the following four types of angle intervals: When 0° ≤ θ < 30°, it is divided into the strong light interval; when 30° ≤ θ < 60°, it is divided into the medium light interval; when 60° ≤ θ < 80°, it is divided into the backlight interval; when 80° ≤ θ ≤ 90°, it is divided into the shadow interval. Then, use the linear mapping function corresponding to each angle interval to perform partitioned gray-scale enhancement on the initial gray-scale image. Further, for the pixels in the strong light interval, apply a low-slope linear mapping function to avoid overexposure in the high-brightness area; for the pixels in the medium light interval, apply a unit-slope linear mapping function to keep the gray-scale structure of the image unchanged; for the pixels in the backlight interval and the shadow interval, apply a linear mapping function with an increasing slope to enhance the image brightness and edge contrast. Finally, use the image after linear remapping as the optimized image data.

[0087] In some embodiments, gray-scale enhancement is performed on the initial gray-scale image based on the light incidence angle value to obtain optimized image data, which specifically includes the following technical steps: Based on the light incidence angle values corresponding to each pixel point in the initial gray-scale image, determine the enhancement weight corresponding to each pixel point, where the enhancement weight is proportional to the cosine value of the light incidence angle value; perform weighted enhancement processing on the original gray-scale values of each pixel point in the initial gray-scale image based on the enhancement weight to generate optimized image data.

[0088] Specifically, the cosine value of the light incidence angle value can be normalized, and the normalized value is used as the enhancement weight. Then, on the basis of the original gray-scale value, the product of the enhancement weight of the pixel point and its original gray-scale value is superimposed to obtain optimized image data. Further, by determining the enhancement weight of each pixel point based on the cosine value of the light incidence angle value and performing weighted enhancement processing on the original gray-scale values of each pixel point in the initial gray-scale image, the brightness unevenness and shadow interference caused by the influence of light can be effectively suppressed, so that on the basis of not destroying the overall gray-scale distribution structure of the image, the gray-scale contrast of the structural feature regions in the image, such as crack boundaries and contour lines, can be enhanced.

[0089] In some embodiments, terrain grid modeling and curvature calculation are performed on the point cloud data to obtain a terrain curvature distribution map, which specifically includes the following technical steps: construct a local neighborhood corresponding to each data point in the point cloud data, and determine the point cloud density and normal variation value corresponding to the local neighborhood; based on the point cloud density and normal variation value, determine the grid size corresponding to each data point; perform non-uniform grid division on the point cloud data according to the grid size to generate a three-dimensional terrain model; perform curvature calculation on each grid node in the three-dimensional terrain model, and generate a terrain curvature distribution map according to the curvature calculation result.

[0090] Among them, the local neighborhood can represent a three-dimensional space region centered on the target data point in the point cloud data and limited by a spatial distance or a point number threshold, and this region is used to obtain a set of several adjacent data points around the target data point. The normal variation value can represent the angle difference between the normal vectors corresponding to all data points in the local neighborhood and the normal vector of the target data point.

[0091] In the specific implementation process, a local neighborhood can be constructed for each data point in the point cloud data. The local neighborhood is a set of spatially adjacent points extracted within a set search radius centered on the data point. Based on the number of points and the distribution range included in the local neighborhood, the corresponding point cloud density is calculated. Secondly, based on the angular distribution between the normal vectors of each data point in the local neighborhood, the normal change value of the target data point is calculated. According to the point cloud density and the normal change value, a weighted function is used to determine the grid cell size corresponding to the data point. The weighted function outputs a smaller grid size value when the normal change value is larger or the point cloud density is higher to achieve adaptive grid adjustment. According to the grid sizes corresponding to all data points, a non-uniform grid division operation is performed on the entire point cloud data to generate a three-dimensional terrain model composed of triangular grid cells of different sizes.

[0092] In addition, when calculating the curvature, for each triangular grid cell in the three-dimensional terrain model, its corresponding spatial coordinates of the vertices can be determined first, and a local fitting plane can be constructed based on multiple adjacent grid nodes in the adjacent area of the triangular cell. Then, based on the local fitting plane, the principal curvature value and the Gaussian curvature value of the current grid node are calculated, and the principal curvature value and the Gaussian curvature value are mapped to the corresponding grid nodes to form a curvature feature set containing the local geometric change characteristics of each node. Finally, the curvature feature sets of all grid nodes are used to generate a terrain curvature distribution map according to their spatial positions in the three-dimensional terrain model to reflect the spatial geometric characteristics of the entire dam slope area.

[0093] In some embodiments, a risk scoring function is used to determine the risk scoring values corresponding to each grid in the joint feature map, specifically including the following technical steps: normalizing the structural feature data and the terrain curvature in the joint feature map respectively to obtain the structural normalized data and the curvature normalized data; using the risk scoring function constructed by the weighting coefficients to perform weighted calculation on the structural normalized data and the curvature normalized data to obtain the risk scoring values corresponding to each grid.

[0094] In the specific implementation, to identify different surface anomaly types in the second risk area, a risk scoring function can be constructed for respectively evaluating the risk levels of the crack feature and the deformation feature: where, represents the crack risk scoring value at the measurement position ; represents the deformation risk scoring value at the measurement position ; represents the structural normalized data; represents the curvature normalized data; represents the direction consistency data of the angle between the image edge direction and the slope normal direction; , , , are preset weight coefficients and can be adjusted according to the actual measurement scenario. Preferably, The value of is 0.8, The value of is 0.2, The value of is 0.75, The value of is 0.25.

[0095] After obtaining the risk score value, if and , then this position is determined as the crack area. If and , then this position is determined as the deformation area. Other situations are determined as normal areas. Then, the crack area and the deformation area are used as the second risk area. Among them, represents the crack identification threshold, represents the deformation identification threshold. Preferably, The value of is 0.65, The value of is 0.6.

[0096] In some embodiments, the above-mentioned dam slope monitoring method based on radar interferometry and aerial survey data further includes: in response to the second risk area being a surface anomaly area, controlling a survey robot to obtain actual deformation data and environmental data of at least one target position in the second risk area; wherein, the actual deformation data includes any one or more of crack width, crack length, and offset height, and the environmental data includes any one or more of air humidity, shallow moisture content, and temperature data.

[0097] Among them, the surface anomaly area can represent an area where terrain anomalies such as deformation and cracks occur in the target dam slope area. In this embodiment, by using the survey robot to perform positioning measurement operations on the second risk area, on-site verification and refined supplementation of the remote sensing measurement results can be achieved, improving the accuracy and measurement integrity of slope risk identification.

[0098] In some embodiments, the above-mentioned dam slope monitoring method based on radar interferometry and aerial survey data further includes: in response to the second risk area being a deformation area, obtaining historical displacement time series data corresponding to the deformation area; performing multi-time scale deformation prediction based on the historical displacement time series data to obtain the deformation prediction result of the deformation area.

[0099] Among them, the deformation area can represent the area where surface deformations such as sliding, dislocation or collapse occur in the dam slope area. Multi-time scale deformation prediction can represent constructing prediction models at different observation periods based on the historical displacement time series data of the target area to reflect the characteristics of short-term and long-term deformations, and generating a prediction result describing the future deformation trend of the target area through weighted fusion of multi-scale prediction results.

[0100] In some embodiments, multi-time scale deformation prediction is performed based on historical displacement time series data to obtain the deformation prediction result of the deformation area, which specifically includes the following technical steps: The historical displacement time series data is segmented using the first time scale and the second time scale respectively to obtain a first-scale subsequence and a second-scale subsequence, where the first time scale is less than the second time scale; Short-term deformation prediction is performed using the first-scale subsequence to obtain a first prediction result; Long-term deformation prediction is performed using the second-scale subsequence to obtain a second prediction result; The first prediction result and the second prediction result are weighted and fused to obtain the deformation prediction result of the deformation area.

[0101] Among them, the first time scale can represent the time period used to characterize the short-term surface deformation trend, which can be 3 days or 7 days or other suitable time periods. The second time scale can represent the time period used to characterize the long-term surface deformation trend, which can be months, quarters or years or other suitable time periods. Exemplarily, when performing deformation prediction, other suitable prediction models such as long short-term memory networks, autoregressive integrated moving average models, and convolutional neural networks can be used. In this embodiment, by dividing the historical displacement data into two time periods, short-term and long-term, and performing predictions separately and then fusing them, sudden deformations and persistent deformations can be taken into account simultaneously, so as to more accurately predict the future changes in the deformation area and improve the reliability and practicality of monitoring and early warning.

[0102] In addition, in other embodiments of the present disclosure, after coordinate alignment of the image structure feature map and the terrain curvature distribution map to obtain a joint feature map, the joint feature map can be further input into an anomaly recognition model, so as to determine a second risk area according to the anomaly recognition result. Specifically, the joint feature map is subjected to image block division processing, that is, the joint feature map is divided into multiple spatially continuous image blocks, and each image block contains image structure feature data and terrain curvature feature data, which are used as the smallest computing unit for anomaly recognition. Each image block is input into a pre-trained anomaly recognition model. The anomaly recognition model can be constructed based on a deep convolutional neural network and has a multi-scale feature extraction structure, which can extract texture, boundary, and curvature change features at different scales and output corresponding anomaly score values. Based on the comparison between the anomaly score value and a set risk discrimination threshold, if the anomaly score value of a certain image block is greater than the risk discrimination threshold, the area corresponding to the image block is marked as a candidate anomaly area. Boundary merging and noise filtering processing are performed on all candidate anomaly areas to exclude isolated blocks and low-confidence edge areas, and areas with strong continuity and high anomaly score values are retained as effective anomaly areas. The effective anomaly area is output as the second risk area for subsequent crack tracking, deformation measurement, or priority areas for measurement robot dispatching tasks.

[0103] It should be noted that although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be executed in that specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0104] In addition, in this exemplary embodiment, a dam slope monitoring system based on radar interferometry and aerial survey data is also provided. Refer to Figure 5 As shown, the dam slope monitoring system 500 based on radar interferometry and aerial survey data includes: a deformation measurement module 510, a first risk area module 520, an aerial survey data acquisition module 530, a feature extraction module 540, a coordinate alignment module 550, and a second risk area module 560. Among them: The deformation measurement module 510 can be used to obtain radar interferometry data of the target dam slope area at multiple time points, perform multi-temporal registration and phase difference processing on the radar interferometry data to obtain displacement time series data; The first risk area module 520 can be used to determine the displacement change data of each measurement position during a preset observation period based on the displacement time series data, and determine the first risk area according to the displacement change data; The aerial survey data acquisition module 530 can be used to determine an aerial survey path according to the first risk area, and control a flight platform to collect aerial survey data of the first risk area based on the aerial survey path. The aerial survey data includes image data and point cloud data; The feature extraction module 540 can be used to extract features from the image data to obtain an image structure feature map, perform terrain grid modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map; The coordinate alignment module 550 can be used to align the coordinates of the image structure feature map and the terrain curvature distribution map to obtain a combined feature map; The second risk area module 560 can be used to determine the risk score value corresponding to each grid in the combined feature map by using a risk scoring function, and determine the second risk area based on the risk score value.

[0105] The specific details of each module of the above dam slope monitoring system based on radar interferometry and aerial survey data have been described in detail in the corresponding dam slope monitoring method based on radar interferometry and aerial survey data, so they will not be elaborated here.

[0106] It should be noted that although several modules or units of the dam slope monitoring system based on radar interferometry and aerial survey data are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0107] In addition, in the exemplary embodiments of the present disclosure, an electronic device capable of implementing the above dam slope monitoring method based on radar interferometry and aerial survey data is also provided.

[0108] Those skilled in the art can understand that various aspects of the present disclosure can be implemented as a system, method, or program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuitry", "module", or "system" here.

[0109] Next, refer to Figure 6 to describe the electronic device 600 according to the embodiments of the present disclosure. Figure 6 The illustrated electronic device 600 is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present disclosure.

[0110] As Figure 6As shown, the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one of the above-mentioned processing units 610, at least one of the above-mentioned storage units 620, a bus 630 connecting different system components (including the storage unit 620 and the processing unit 610), and a display unit 640.

[0111] Among them, the storage unit stores program code, and the program code can be executed by the processing unit 610, so that the processing unit 610 executes the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section of this specification.

[0112] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 621 and / or a cache storage unit 622, and may further include a read-only storage unit (ROM) 623.

[0113] The storage unit 620 may also include a program / utilities 624 having a set (at least one) of program modules 625. Such program modules 625 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0114] The bus 630 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus structures.

[0115] The electronic device 600 can also communicate with one or more external devices 670 (such as a keyboard, a pointing device, a Bluetooth device, etc.), can also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or can communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication can be carried out through an input / output (I / O) interface 650. And, the electronic device 600 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 660. As shown in the figure, the network adapter 660 communicates with other modules of the electronic device 600 through the bus 630. It should be understood that although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0116] From the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0117] In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium having a program product stored thereon that can implement the methods described above in this specification. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program code that, when the program product runs on a terminal device, causes the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Methods" section above in this specification.

[0118] Reference Figure 7 As shown, a program product 700 for implementing the above-mentioned dam slope monitoring method based on radar interferometry and aerial survey data according to an embodiment of the present disclosure is described. It can be a portable compact disc read-only memory and includes program code and can run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0119] The program product can adopt any combination of one or more readable media. A readable medium can be a readable signal medium or a readable storage medium. A readable storage medium can, for example, but not be limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disc read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0120] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable signal medium may also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0121] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0122] The program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computing device, partially on the user's device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device through any type of network, including a local area network or a wide area network, or may be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0123] In addition, the above drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not for limiting purposes. It is easy to understand that the processes shown in the above drawings do not indicate or limit the chronological order of these processes. Additionally, it is also easy to understand that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0124] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which may be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which may be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0125] Other embodiments of the present disclosure will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known or customary technical means in the art not disclosed herein. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.

[0126] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A dam slope monitoring method based on radar interferometry and aerial survey data, characterized in that: include: Acquire radar interferometry data of the target dam slope area at multiple time points, perform multi-temporal phase registration and phase difference processing on the radar interferometry data, and obtain displacement time series data; Determine displacement change data of each measurement position in a preset observation period based on the displacement time series data, and determine a first risk area according to the displacement change data; Determine an aerial survey path according to the first risk area, and control the flight platform to collect aerial survey data of the first risk area based on the aerial survey path, wherein the aerial survey data includes image data and point cloud data; Performing feature extraction on the image data to obtain an image structure feature map, and performing terrain gridding modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map; Aligning the image structure feature map with the terrain curvature distribution map to obtain a joint feature map; A risk score function is used to determine a risk score value corresponding to each grid in the joint feature map, and a second risk area is determined based on the risk score value.

2. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1 is characterized in that: The radar interferometry data is subjected to multi-phase registration and phase difference processing to obtain displacement time series data, including: Based on orbital parameters and geographic reference coordinates, image registration processing is performed on the radar interferometry data to obtain a multi-temporal radar image set; Performing coherence screening and phase difference calculation on the multi-phase radar image set to obtain phase time series data corresponding to a plurality of phase stable points; Based on the position coordinates of the phase stable point and the phase time series data, the displacement time series data for describing the surface deformation is constructed.

3. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 2 is characterized in that: The coherence screening and phase difference calculation are performed on the multi-phase radar image set to obtain phase time series data corresponding to multiple phase stable points, including: Determine an interference phase difference between adjacent time points for each image pixel in the multi-temporal radar image set, and determine a coherence index value of the corresponding image pixel based on the interference phase difference; Comparing the coherence index value with a preset coherence threshold, and selecting a plurality of phase stable points from the image pixels according to the comparison result; The phase time series data is constructed based on the interference phase difference corresponding to the phase stable point.

4. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1 is characterized in that: The step of determining displacement change data of each measurement position in a preset observation period based on the displacement time series data, and determining a first risk area according to the displacement change data, comprises: Determining the displacement amount and displacement rate of each measurement position in the displacement time series data in a preset observation period; Based on the displacement amount and the displacement rate, screening out a first risk position from the measurement positions; The first risk area is generated based on the location distribution of the first risk locations.

5. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1 is characterized in that: The determining of the aerial survey path according to the first risk area includes: Converting the image pixel coordinates corresponding to the measurement position in the first risk area into geographic space coordinates through coordinate solution and space projection processing; At least one of the aerial survey paths is determined based on the geospatial coordinates.

6. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1 is characterized in that: The step of extracting features from the image data to obtain an image structure feature map comprises: Performing grayscale conversion on the image data to obtain an initial grayscale image; Using the three-dimensional terrain model corresponding to the image data, the initial grayscale image is subjected to grayscale enhancement processing to obtain optimized image data; Edge feature extraction and grayscale change feature extraction are performed on the optimized image data, and the image structure feature map is generated based on the feature extraction results.

7. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 6 is characterized in that: The method of performing grayscale enhancement processing on the initial grayscale image using the three-dimensional terrain model corresponding to the image data to obtain optimized image data includes: Constructing a three-dimensional terrain model based on the point cloud data of the area corresponding to the image data, and determining the normal direction of the slope corresponding to each pixel point in the initial grayscale image based on the three-dimensional terrain model; Determining a light incident direction vector corresponding to the image data according to the acquisition time and geographical location of the image data; Angle calculation is performed based on the normal direction of the slope surface and the light incident direction vector to generate a light incident angle value corresponding to each pixel point; The initial grayscale image is enhanced in grayscale based on the light incident angle value to obtain the optimized image data.

8. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 7 is characterized in that: The step of performing grayscale enhancement on the initial grayscale image based on the light incident angle value to obtain the optimized image data includes: Dividing the illumination incident angle value corresponding to each pixel point in the initial grayscale image into corresponding angle intervals, wherein the angle intervals include a bright light interval, a medium light interval, a backlight interval, and a shadow interval; Using the grayscale remapping function corresponding to each of the angle intervals, performing a partitioned grayscale enhancement process on the initial grayscale image to obtain the optimized image data; The grayscale remapping function includes any one of a linear mapping function, a gamma correction function and a logarithmic enhancement function.

9. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 7 is characterized in that: The step of performing grayscale enhancement on the initial grayscale image based on the light incident angle value to obtain the optimized image data includes: Based on the light incident angle value corresponding to each pixel in the initial grayscale image, determining an enhancement weight corresponding to each pixel, wherein the enhancement weight is proportional to the cosine value of the light incident angle value; The original grayscale value of each pixel in the initial grayscale image is subjected to weighted enhancement processing based on the enhancement weight to generate the optimized image data.

10. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1, characterized in that: The performing terrain gridding modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map includes: Constructing a local neighborhood corresponding to each data point in the point cloud data, and determining a point cloud density and a normal change value corresponding to the local neighborhood; Determine a grid size corresponding to each of the data points based on the point cloud density and the normal change value; Performing non-uniform grid division on the point cloud data according to the grid size to generate a three-dimensional terrain model; The curvature of each grid node in the three-dimensional terrain model is calculated, and the terrain curvature distribution map is generated according to the curvature calculation result.

11. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1, characterized in that: Determining the risk score value corresponding to each grid in the joint feature graph by using the risk score function includes: Normalizing the structural feature data and the terrain curvature in the joint feature map respectively to obtain structural normalized data and curvature normalized data; The risk score function constructed by weighting coefficients is used to perform weighted calculation on the structure normalized data and the curvature normalized data to obtain the risk score value corresponding to each grid.

12. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1, characterized in that: Also includes: In response to the second risk area being a surface abnormal area, controlling the measurement robot to obtain actual deformation data and environmental data of at least one target position in the second risk area; The actual deformation data include any one or more of crack width, crack length and misalignment height, and the environmental data include any one or more of air humidity, shallow layer moisture content and temperature data.

13. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 1, characterized in that: Also includes: In response to the second risk area being a deformation area, acquiring historical displacement time series data corresponding to the deformation area; Multi-time-scale deformation prediction is performed based on the historical displacement time series data to obtain a deformation prediction result of the deformation area.

14. The dam slope monitoring method based on radar interferometry and aerial survey data according to claim 13, characterized in that: The multi-time-scale deformation prediction is performed based on the historical displacement time series data to obtain the deformation prediction result of the deformation area, including: The historical displacement time series data is segmented using a first time scale and a second time scale to obtain a first scale subsequence and a second scale subsequence, wherein the first time scale is smaller than the second time scale; Using the first scale subsequence to perform short-term deformation prediction to obtain a first prediction result; Using the second scale subsequence to perform long-term deformation prediction to obtain a second prediction result; The first prediction result and the second prediction result are weightedly fused to obtain a deformation prediction result of the deformation area.

15. A dam slope monitoring system based on radar interferometry and aerial survey data, characterized in that: include: The deformation measurement module is used to obtain radar interferometry data of the target dam slope area at multiple time points, perform multi-time phase registration and phase difference processing on the radar interferometry data, and obtain displacement time series data; A first risk area module, configured to determine displacement change data of each measurement position in a preset observation period based on the displacement time series data, and determine a first risk area according to the displacement change data; an aerial survey data acquisition module, configured to determine an aerial survey path according to the first risk area, and control the flight platform to collect aerial survey data of the first risk area based on the aerial survey path, wherein the aerial survey data includes image data and point cloud data; A feature extraction module is used to extract features from the image data to obtain an image structure feature map, and to perform terrain gridding modeling and curvature calculation on the point cloud data to obtain a terrain curvature distribution map; A coordinate alignment module, used for performing coordinate alignment on the image structure feature map and the terrain curvature distribution map to obtain a joint feature map; The second risk area module is used to determine the risk score value corresponding to each grid in the joint feature map by using a risk score function, and determine the second risk area based on the risk score value.

Images