A High-Frequency DEM Update Method and Equipment Based on Repeating Orbit InSAR

By acquiring repeating orbit SAR image data, unwrapping the phase matrix using DInSAR technology and performing robust estimation, the problem of low accuracy of short-time DEM updates under repeating orbit InSAR technology is solved, and high-frequency DEM updates are achieved.

CN117473215BActive Publication Date: 2026-11-17CENT SOUTH UNIV +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311383585.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-10-24
Publication Date
2026-11-17
Estimated Expiration
2043-10-24

AI Technical Summary

Technical Problem

In highly disturbed areas, the accuracy of DEM updates under existing heavy-orbit InSAR technology is low in short time series, making it difficult to achieve high-frequency updates.

Method used

By acquiring repeating orbit SAR image data of the target area at multiple times, the phase matrix is ​​unwrapped using DInSAR technology, an observation equation set is constructed, robust estimation and weight matrix update are performed, the terrain residual error is reduced, and the accuracy of DEM update is improved.

Benefits of technology

It enables high-frequency updates of DEM under short time series, improving the accuracy and timeliness of DEM updates, and is suitable for DEM updates in highly disturbed areas.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117473215B_ABST
    Figure CN117473215B_ABST
Patent Text Reader

Abstract

This application relates to the field of acquiring digital elevation models (DEMs) of the Earth's surface, and provides a method and apparatus for high-frequency updating of DEMs based on repeating orbit InSAR. The method includes: acquiring multiple repeating orbit SAR image data; unwrapping the data to obtain an unwrapped phase matrix; constructing an observation equation set based on the multi-orbit unwrapped phase matrix and obtaining an initial solution to the observation equation set; performing robust estimation on the initial solution to obtain a robust weight matrix; calculating the posterior variance matrix; updating the robust weight matrix using the posterior variance matrix to obtain a weight matrix; determining whether the elements in the posterior variance matrix are equal; if so, the weight matrix is ​​used as the optimal weight matrix; otherwise, the weight matrix is ​​used as the initial weight matrix, and the process returns to the step of robustly estimating the initial solution to obtain the robust weight matrix; estimating the terrain residual at time T using the optimal weight matrix, and adding it to the reference DEM to obtain the current DEM. This method can solve the problem of low accuracy in DEM updates.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of terrain monitoring technology, and in particular to a high-frequency update method and device for DEM based on repeating orbit InSAR. Background Technology

[0002] Digital elevation models (DEMs) play a crucial role in numerous practical applications, including civil engineering, urban planning, topographic mapping, and landslide monitoring. For applications in highly disturbed areas, the timeliness of the DEM is paramount. For instance, when monitoring mining output in open-pit mines, updating the DEM is necessary to obtain the ore extraction volume for the corresponding time period. Currently, the main technologies used to generate large-scale DEMs include optical photogrammetry and interferometric synthetic aperture radar (InSAR). However, optical photogrammetry is susceptible to weather conditions such as rain and fog, making it difficult to update DEM products frequently. In contrast, InSAR technology, with its unique historical backtracking, all-weather, all-time capabilities, and high spatiotemporal resolution, offers significant advantages in DEM production and updating.

[0003] Depending on the satellite imaging mode, InSAR methods for generating DEMs can be divided into two categories: 1) Single-track dual-antenna mode, such as the TanDEM-X DEM acquired using the bistatic mode of the TerraSAR-X and TanDEM-X dual-satellite formation flight, and the Shuttle Radar Topography Mission (SRTM); 2) Heavy orbit interferometry mode. For the single-track dual-antenna mode, the acquired interferometric phase is less affected by atmospheric and temporal decoherence, resulting in higher accuracy DEM products. However, the satellite revisit period in this mode is long, making it difficult to guarantee the timeliness of the DEM products. In contrast, the heavy orbit mode has a shorter satellite revisit period (Sentinel-1 satellite can provide Synthetic Aperture Radar (SAR) images at 6-day intervals) and more available data. Furthermore, with the continued development of heavy orbit satellites in the future, the amount of data in this type will become increasingly abundant, ensuring the timeliness of DEM products. Therefore, if the potential of updating DEMs using heavy orbit InSAR data can be fully explored, it may be possible to achieve high-frequency updates of DEM products over a wide range.

[0004] However, the terrain phase acquired in the double-track mode is easily affected by atmospheric and surface deformation factors. Some scholars have estimated the deformation phase and DEM residual phase from the interferometric phase using time-series InSAR technology. The external DEM can then be updated and corrected using the elevation residual information. Subsequently, to improve the DEM residual correction capability, scholars at home and abroad have conducted a lot of research based on the traditional differential interferometry short baseline subset InSAR (SBAS-InSAR) technology. However, such methods usually set the terrain parameters as constants, and current studies almost all use radar observation data acquired over a long period of time for calculation. In high-disturbance scenarios, the timeliness of the DEM reconstructed in long time series is very poor, while the terrain parameters estimated in short time series are easily affected by noise due to fewer redundant observations, resulting in low accuracy of DEM updates in short time series. Summary of the Invention

[0005] This application provides a method and device for high-frequency DEM updating based on repeating orbit InSAR, which can solve the problem of low accuracy of DEM updating under short time series.

[0006] In a first aspect, embodiments of this application provide a high-frequency DEM update method based on repeating orbit InSAR, which includes:

[0007] Acquire SAR image data of the target area at multiple repeating orbits at time T; the SAR image data includes SAR images of multiple observation phases, with time T being the current time;

[0008] For each repeating orbit, the SAR image data is unwrapped using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit.

[0009] The observation equations are constructed based on the unwrapped phase matrix of all repeating orbits, and the initial solutions of the topographic residuals and surface deformation parameters in the observation equations are obtained.

[0010] A robust weight matrix is ​​obtained by performing a robust estimation on the initial weight matrix in the initial solution.

[0011] The posterior variance matrix corresponding to the target region is calculated using the initial solution and the robust weight matrix, and the robust weight matrix is ​​updated using the posterior variance matrix to obtain the weight matrix.

[0012] Determine whether the elements in the posterior variance matrix are equal. If they are, use the weight matrix as the optimal weight matrix. Otherwise, use the weight matrix as the initial weight matrix in the initial solution and return to the step of performing robust estimation on the initial weight matrix in the initial solution to obtain the robust weight matrix.

[0013] The observation equations are updated using the optimal weight matrix, and the updated observation equations are used to estimate the terrain residuals of the target area at time T.

[0014] The topographic residual of the target area at time T is added to the external reference DEM of the target area to obtain the DEM of the target area at the current time.

[0015] Optionally, the initial solutions for the topographic residuals and surface deformation parameters are:

[0016]

[0017] P = [P1,...,P] i ,...,P n ]

[0018] in, Let P represent the topographic residuals and surface deformation parameters, P1 represent the initial weight matrix of the initial solution of the topographic residuals and surface deformation parameters, and P1 represent the initial observation phase weight matrix of the first repeated orbit. i P represents the initial observation phase weight matrix for the i-th repeating orbit. n Let represent the initial observation phase weight matrix for the nth repeating orbit, i = 1, 2, ..., n, where n represents the total number of repeating orbits. T This represents the matrix transpose operation. Represents the parameter matrix, B represents the transformation matrix between parameters and unwrapped phase. geo Represents the projection transformation matrix. This represents the unwrapping phase matrix for all repeating orbits.

[0019] Optionally, a robust weight matrix is ​​obtained by robustly estimating the initial weight matrix in the initial solution, including:

[0020] Through the formula:

[0021]

[0022] Calculate the observation residual vector matrix V for the i-th repeating orbit. i ;

[0023] Among them, V i =[V i,1 ,...,V i,r ,...,V i,R ], V i,1 V represents the observation residual vector of the first observation phase of the i-th repeating orbit. i,rV represents the observation residual vector of the r-th observation phase of the i-th repeating orbit. i,R Let r represent the observation residual vector of the R-th observation phase of the i-th repeating orbit, where r = 1, 2, ..., R, and R represents the total number of observation phases. This represents the parameter matrix of the i-th repeating orbit. This represents the initial solution for topographic residuals and surface deformation parameters. Let represent the unwrapping phase matrix of the i-th repeating orbit;

[0024] Through the formula:

[0025]

[0026]

[0027] Calculate the normalized observation residual vector matrix of the i-th repeating orbit.

[0028] Where ξ represents the robust estimation parameter, and median() represents the calculation of the median. This represents the standard observation residual of the first observation phase of the i-th repeating orbit. This represents the standard observation residual of the r-th observation phase of the i-th repeating orbit. This represents the standard observation residual of the R-th observation phase of the i-th repeating orbit;

[0029] Through the formula:

[0030]

[0031] Calculate the robust phase weight of the r-th observation phase of the i-th repeating orbit.

[0032] Among them, P i,r P represents the initial observation phase weight of the r-th observation phase of the i-th repeating orbit. i,r ∈P i Both k0 and k1 are constants, and k0 <k1;

[0033] Obtaining a robust weight matrix

[0034] in, This represents the robust observation phase weight matrix for the first repeating orbit. The robust observation phase weight represents the first observation phase of the first repeating orbit. This represents the robust observation phase weight for the r-th observation phase of the first repeating orbit. This represents the robust observation phase weight for the R-th observation phase of the first repeating orbit. This represents the robust observation phase weight matrix for the i-th repeating orbit. This represents the robust observation phase weight of the first observation phase of the i-th repeating orbit. This represents the robust observation phase weight for the R-th observation phase of the i-th repeating orbit. This represents the robust observation phase weight matrix for the nth repeating orbit. This represents the robust observation phase weight of the first observation phase of the nth repeating orbit. This represents the robust observation phase weight for the r-th observation phase of the n-th repeating orbit. This represents the robust observation phase weight of the Rth observation phase of the nth repeating orbit.

[0035] Optionally, the posterior variance matrix corresponding to the target region is calculated using the initial solution and the robust weight matrix, including:

[0036] Through the formula:

[0037]

[0038] Calculate the posterior variance matrix corresponding to the target region.

[0039] in, This represents the posterior variance of the observation residual vector for the first repeated orbit. This represents the posterior variance of the observation residual vector for the second repeated orbit. Let Vn represent the posterior variance of the observation residual vector of the nth repeating orbit, and V1 represent the observation residual vector matrix of the 1st repeating orbit. n Let s represent the observation residual vector matrix of the nth repeating orbit, j = 1, 2, ..., n. 1,1 s 1,n s i,1 s i,i s i,n s n,1 and s n,n All represent posterior variance parameters:

[0040] s i,i =n i -2tr(N -1 N i )+tr(N -1 N i N -1 N i )

[0041] s i,j =n i -tr(N -1 N i N -1 N j )

[0042]

[0043] N = N1 + N2 + ... + N n

[0044] Where, n i N represents i The i-th matrix element in the matrix, tr() represents the trace of the matrix, N represents the coefficient matrix of the normal equation, N i N represents the normal equation coefficient matrix corresponding to the i-th repeating orbit. j This represents the normal equation coefficient matrix corresponding to the j-th repeating orbit. Ni represents the coefficient matrix corresponding to the i-th repeating orbit, N1 represents the normal equation coefficient matrix corresponding to the 1st repeating orbit, N2 represents the normal equation coefficient matrix corresponding to the 2nd repeating orbit, and Ni represents the normal equation coefficient matrix corresponding to the i-th repeating orbit. n Let N = [N1, ..., N] represent the normal equation coefficient matrix corresponding to the nth repeating orbit. n ].

[0045] Optionally, the robust weight matrix can be updated using the posterior variance matrix to obtain the weight matrix, including:

[0046] Through the formula:

[0047]

[0048] The elements of the initial weight matrix are updated to obtain the final weight matrix.

[0049] in, This represents the observation phase weight matrix for the first repeating orbit. This represents the observation phase weight matrix for the i-th repeating orbit. This represents the observation phase weight matrix for the nth repeating orbit.

[0050] Optionally, the updated observation equations are used to estimate the topographic residuals of the target region at time T, including:

[0051] Through the formula:

[0052]

[0053] The deformation parameter vector matrix is ​​calculated. and the terrain residual Δh at time T;

[0054] in, Represents the parameter matrix, B represents the transformation matrix between parameters and unwrapped phase. geo Let P represent the projection transformation matrix. * This represents the optimal weight matrix. Represents the unwrapped phase matrix of all repeating orbits, () T This represents the matrix transpose operation.

[0055] Optionally, the SAR image data can be unwrapped using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit, including:

[0056] Multiple interferometric pairs in SAR image data are generated using DInSAR technology;

[0057] Using the external reference DEM of the target area, differential processing is performed on each interference pair to obtain differential interference pairs;

[0058] The differential interferometer pair is unwrapped to obtain the unwrapped phase, and the unwrapped phase matrix of the repeating orbit is obtained.

[0059] Optionally, a set of observation equations is constructed based on the unwrapped phase matrix of all repeating orbits, including:

[0060] Based on the unwrapped phase matrix of all repeating orbits, an observation equation is constructed between the phase and the topographic residuals and surface deformation parameters:

[0061]

[0062] Where B1 represents the coefficient matrix of the first repeating orbit, B n Let (M′×13), ((4n+1)×1), and (M′×1) represent the coefficient matrix of the nth repeating orbit, where (M′×13), ((4n+1)×1), and (M′×1) all represent the matrix dimensions, and M′ represents the total number of differential interference pairs of the n repeating orbits. Represents the deformation parameter vector matrix. This represents the deformation parameter vector of the first repeating orbit. This represents the deformation parameter vector of the second repeating orbit. Let [ ] represent the deformation parameter vector of the nth repeating orbit. T The transpose operation of a matrix is ​​represented by Δ. (M′×1) Indicates the residual phase. This represents the unwrapping phase matrix of the first repeating orbit. This represents the unwrapping phase of the first differential interference pair of the first repeating orbit. This represents the unwrapping phase of the second differential interference pair of the first repeating orbit. This represents the unwrapping phase of the Lth differential interference pair of the first repeating orbit. This represents the unwrapping phase matrix of the Nth repeating orbit. This represents the unwrapping phase of the first differential interference pair of the Nth repeating orbit. This represents the unwrapping phase of the second differential interference pair of the Nth repeating orbit. This represents the unwrapping phase of the Lth differential interference pair of the Nth repeating orbit;

[0063] For each of the multiple repeating orbits, the radar line-of-sight deformation parameter vector corresponding to the repeating orbit is projected onto the east-west and vertical directions, resulting in the following set of observation equations:

[0064]

[0065] Where (M′×(2n+1)), ((2N+1)×5), and (5×1) all represent the matrix dimensions. Represents topographic residuals and surface deformation parameters. This represents the average rate of deformation in the vertical direction. This represents the average rate of deformation along the east-west direction. This represents the vertical deformation acceleration. Indicates the deformation acceleration in the east-west direction. This represents the unwrapping phase matrix for all repeating orbits. B represents the transformation matrix between parameters and unwrapped phase. geo Representing the projection transformation matrix:

[0066]

[0067]

[0068] Where, θ 1 θ represents the angle of incidence of the satellite in the first repeating orbit. 2 α represents the angle of incidence of the satellite in the second repeating orbit. 1 α represents the azimuth angle of the first repeating orbit satellite. 2 Cosθ represents the azimuth angle of the second repeating orbit satellite. 1 Cosθ represents the projection relationship between the slant range direction of the satellite and the vertical direction to the Earth's surface on the first repeating orbit. 2-sinθ represents the projection relationship between the slant range direction of the satellite and the vertical direction to the Earth's surface on the second repeating orbit. 1 cosα 1 -sinθ represents the projection relationship between the slant range of the satellite in the first repeating orbit and the east-west direction of the Earth's surface. 2 csoα 2 This describes the projection relationship between the slant range of the satellite in the second repeating orbit and the east-west direction of the Earth's surface.

[0069] Secondly, embodiments of this application provide a DEM high-frequency update device based on repeating orbit InSAR, comprising:

[0070] The acquisition module acquires SAR image data of the target area at T time points using multiple repeating orbits; the SAR image data includes SAR images from multiple observation phases, with the Tth time point being the current time.

[0071] The unwrapping module is used to unwrap the SAR image data for each repeating orbit using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit.

[0072] The module is used to construct the observation equation set based on the unwrapped phase matrix of all repeating orbits, and to obtain the initial solutions of the terrain residual and surface deformation parameters in the observation equation set.

[0073] The calculation module is used to robustly estimate the initial weight matrix in the initial solution to obtain the robust weight matrix;

[0074] The update module calculates the posterior variance matrix corresponding to the target region using the initial solution and the robust weight matrix, and updates the robust weight matrix using the posterior variance matrix to obtain the weight matrix.

[0075] The judgment module determines whether the elements in the posterior variance matrix are equal. If they are equal, the weight matrix is ​​used as the optimal weight matrix. Otherwise, the weight matrix is ​​used as the initial weight matrix in the initial solution, and the module returns the step of performing robust estimation on the initial weight matrix in the initial solution to obtain the robust weight matrix.

[0076] The estimation module updates the observation equations using the optimal weight matrix and uses the updated observation equations to estimate the terrain residuals of the target area at time T.

[0077] The addition module is used to add the terrain residual of the target area at time T to the external reference DEM of the target area to obtain the DEM of the target area at the current time.

[0078] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the aforementioned high-frequency DEM update method based on repeating orbit InSAR.

[0079] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned high-frequency DEM update method based on repeating orbit InSAR.

[0080] The above-mentioned solution in this application has the following beneficial effects:

[0081] In the embodiments of this application, SAR image data of multiple repeating orbits of the target area at T time points are acquired. Then, for each repeating orbit, the SAR image data is unwrapped using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit. Based on the unwrapped phase matrices of all repeating orbits, an observation equation set is constructed, which increases the amount of data contained in the observation equation set and thus improves its accuracy. Then, the initial solutions for the terrain residuals and surface deformation parameters in the observation equation set are obtained. The initial weight matrix in the initial solution is robustly estimated to obtain a robust weight matrix. The posterior variance matrix corresponding to the target area is calculated using the initial solution and the robust weight matrix, and the robust weight matrix is ​​updated using the posterior variance matrix to obtain the weight matrix. Finally, a judgment is made. If the elements in the posterior variance matrix are equal, the weight matrix is ​​used as the optimal weight matrix; otherwise, it is used as the initial weight matrix in the initial solution. The process then returns to the step of robustly estimating the initial weight matrix in the initial solution to obtain a robust weight matrix. By performing robust estimation and using the posterior variance matrix, the error of the optimal weight matrix can be reduced and its accuracy improved. The optimal weight matrix is ​​then used to update the observation equations, and the updated observation equations are used to estimate the terrain residual of the target area at time T. This significantly reduces the error of the terrain residual and improves the accuracy. Finally, when the terrain residual of the target area at time T is added to the external reference DEM of the target area to obtain the DEM of the target area at the current time, the accuracy of the DEM update is effectively improved.

[0082] Furthermore, this application acquires SAR image data from multiple repeating orbits, and the SAR image data includes SAR images from multiple observation phases at multiple times. This decomposes the terrain deformation of the target area over a period of time into terrain at multiple times, which is suitable for DEM updates in short time series and ensures the amount of data in short time series, thus achieving accurate DEM updates in short time series.

[0083] Other beneficial effects of this application will be described in detail in the following detailed description section. Attached Figure Description

[0084] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0085] Figure 1 A flowchart illustrating a high-frequency DEM update method based on repeating orbit InSAR provided in an embodiment of this application;

[0086] Figure 2 An error histogram of the updated DEM provided in one embodiment of this application;

[0087] Figure 3 A schematic diagram of the structure of a high-frequency DEM update device based on repeating orbit InSAR provided in an embodiment of this application;

[0088] Figure 4 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0089] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0090] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0091] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0092] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0093] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0094] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0095] To address the low accuracy of existing short-time-series DEM updates, this application provides a high-frequency DEM update method based on repeating orbit InSAR. This method acquires SAR image data of the target area at T time points using multiple repeating orbits. Then, for each repeating orbit, the SAR image data is unwrapped using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit. An observation equation set is then constructed based on the unwrapped phase matrices of all repeating orbits, increasing the amount of data contained in the observation equation set and thus improving its accuracy. Initial solutions for topographic residuals and surface deformation parameters in the observation equation set are then obtained. Robust estimation of the initial weight matrix in the initial solution yields a robust weight matrix. Finally, the posterior equation corresponding to the target area is calculated using the initial solution and the robust weight matrix. The system calculates the difference matrix and updates the robust weight matrix using the posterior variance matrix. It then checks if the elements in the posterior variance matrix are equal; if so, the weight matrix is ​​used as the optimal weight matrix; otherwise, it is used as the initial weight matrix in the initial solution. The process then returns to the step of robustly estimating the initial weight matrix in the initial solution to obtain the robust weight matrix. Using robust estimation and the posterior variance matrix reduces the error of the optimal weight matrix and improves its accuracy. The optimal weight matrix is ​​then used to update the observation equations, and the updated observation equations are used to estimate the terrain residual of the target area at time T. This significantly reduces the error of the terrain residual and improves accuracy. Finally, when the terrain residual of the target area at time T is added to the external reference DEM of the target area to obtain the DEM of the target area at the current time, the accuracy of the DEM update is effectively improved.

[0096] Furthermore, this application acquires SAR image data from multiple repeating orbits, and the SAR image data includes SAR images from multiple observation phases at multiple times. This decomposes the terrain deformation of the target area over a period of time into terrain at multiple times, which is suitable for DEM updates in short time series and ensures the amount of data in short time series, thus achieving accurate DEM updates in short time series.

[0097] The following section provides an exemplary description of the high-frequency DEM update method based on repeating orbit InSAR of this application.

[0098] like Figure 1 As shown, the high-frequency DEM update method based on repeating orbit InSAR of this application includes the following steps:

[0099] Step 11: Obtain SAR image data of the target area at T time points using multiple repeating orbits.

[0100] The aforementioned SAR image data includes SAR images from multiple observation phases, with the Tth time being the current time. The aforementioned repeating orbits are satellite orbits that coincide with other satellite orbits over a period of time, and the aforementioned target area is the area that needs to be updated in the DEM.

[0101] In some embodiments of this application, the SAR image data may be SAR images acquired by Sentinel-1 series A / B satellites (synthetic aperture radar observation satellites designed and launched by the European Space Agency, which are capable of acquiring image data with short revisit periods, good interferometric quality and accurate orbital attitude).

[0102] It is worth mentioning that the acquisition of SAR image data from multiple repeating orbits, including SAR images from multiple observation phases at multiple times, ensures a large amount of data in a short time series. At the same time, since the terrain deformation of the target area is not uniform, the terrain deformation of the target area over a period of time is decomposed into multiple time periods, which facilitates the study of dynamic terrain deformation.

[0103] Step 12: For each repeating orbit, the SAR image data is unwrapped using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit.

[0104] In some embodiments of this application, the steps of unwrapping the SAR image data using DInSAR technology for each repeating orbit and obtaining the unwrapped phase matrix of the repeating orbit are specifically as follows:

[0105] The first step is to use DInSAR technology to generate multiple interferometric pairs in the SAR image data.

[0106] The second step involves using the external reference DEM of the target area to perform differential processing on each interference pair to obtain differential interference pairs.

[0107] The third step is to unwrap the differential interference pair to obtain the unwrapped phase and acquire the unwrapped phase matrix of the repeating orbit.

[0108] For example, in Differential InSAR (DInSAR), SAR images acquired from different repeating orbits are first preprocessed to generate Single Look Complex (SLC) images. Then, SLC images are registered, and the temporal and spatial baselines between each main and secondary image are calculated based on the image orbit information and parameters in the header file. Appropriate spatiotemporal baseline thresholds are set to generate differential interferometric pairs. Each differential interferometric pair is then unwrapped, specifically through multi-look processing, adaptive filtering, phase unwrapping, polynomial fitting (removing residual orbital phase), and spatial domain filtering (removing atmospheric phase).

[0109] In some embodiments of this application, when performing multi-view processing, to avoid DEM distortion due to low resolution of the 3D surface model in steep terrain application scenarios, the multi-view ratio (azimuth: slant range) can be set to 1:4. When performing phase unwrapping, the selected unwrapping starting point should be located at a stable point on the surface, and the distance between it and the target area should be less than a preset distance value; otherwise, unwrapping failure is likely.

[0110] It should be noted that the external reference DEM of the target area mentioned above is the existing DEM of the target area, and this DEM has not been updated. The differential interferometric pair mentioned above consists of two SAR images from the SAR image data. After step 12 is completed, it is also necessary to use the external reference DEM of the target area and the parameter information of the repeating orbit to geocode the unwrapped phase of the differential interferometric pair so that the unwrapped phase of the differential interferometric pair is located in the geographic coordinate grid sequence.

[0111] It is worth mentioning that unwrapping SAR image data with repeating orbits can reveal the topographic changes in the target area and obtain the unwrapped phase, which is convenient for subsequent calculations.

[0112] Step 13: Construct the observation equation set based on the unwrapped phase matrix of all repeating orbits, and obtain the initial solutions of the terrain residual and surface deformation parameters in the observation equation set.

[0113] In some embodiments of this application, the steps of constructing the observation equation set based on the unwrapped phase matrix of all repeating orbits and obtaining the initial solutions of the terrain residuals and surface deformation parameters in the observation equation set specifically include:

[0114] The first step is to construct the observation equations between phase, topographic residuals, and surface deformation parameters based on the unwrapped phase matrix of all repeating orbits:

[0115]

[0116] Where B1 represents the coefficient matrix of the first repeating orbit, Bn Let (M′×13), ((4n+1)×1), and (M′×1) represent the coefficient matrix of the nth repeating orbit, where (M′×13), ((4n+1)×1), and (M′×1) all represent the matrix dimensions, and M′ represents the total number of differential interference pairs of the n repeating orbits. Represents the deformation parameter vector matrix. This represents the deformation parameter vector of the first repeating orbit. This represents the deformation parameter vector of the second repeating orbit. Let [ ] represent the deformation parameter vector of the nth repeating orbit. T The transpose operation of a matrix is ​​represented by Δ. (M′×1) Indicates the residual phase. This represents the unwrapping phase matrix of the first repeating orbit. This indicates the unwrapping phase of the first interference pair in the first repeating orbit. This indicates the unwrapping phase of the second interference pair in the first repeating orbit. This represents the unwrapping phase of the Lth interference pair of the first repeating orbit. This represents the unwrapping phase matrix of the nth repeating orbit. This represents the unwrapping phase of the first interference pair in the nth repeating orbit. This represents the unwrapping phase of the second interference pair in the nth repeating orbit. This represents the unwrapping phase of the Lth interference pair of the nth repeating orbit.

[0117] The second step involves projecting the radar line-of-sight deformation parameter vector corresponding to each of the multiple repeating orbits onto the east-west and vertical directions, resulting in the following set of observation equations:

[0118]

[0119] Where (M′×(2n+1)), ((2n+1)×5), and (5×1) all represent the matrix dimensions. Represents topographic residuals and surface deformation parameters. This represents the average rate of deformation in the vertical direction. This represents the average rate of deformation along the east-west direction. This represents the vertical deformation acceleration. Indicates the deformation acceleration in the east-west direction. This represents the unwrapping phase matrix for all repeating orbits. B represents the transformation matrix between parameters and unwrapped phase. geo Representing the projection transformation matrix:

[0120]

[0121] Where, θ 1 θ represents the angle of incidence of the satellite in the first repeating orbit. 2 α represents the angle of incidence of the satellite in the second repeating orbit. 1 α represents the azimuth angle of the first repeating orbit satellite. 2 Cosθ represents the azimuth angle of the second repeating orbit satellite. 1 Cosθ represents the projection relationship between the slant range direction of the satellite and the vertical direction to the Earth's surface on the first repeating orbit. 2 -sinθ represents the projection relationship between the slant range direction of the satellite and the vertical direction to the Earth's surface on the second repeating orbit. 1 cosα 1 -sinθ represents the projection relationship between the slant range of the satellite in the first repeating orbit and the east-west direction of the Earth's surface. 2 cosα 2 This describes the projection relationship between the slant range of the satellite in the second repeating orbit and the east-west direction of the Earth's surface.

[0122] The third step is to use least squares to solve the observation equations to obtain the initial solutions for the topographic residuals and surface deformation parameters:

[0123]

[0124] P = [P1,...,P] i ,...,P n ]

[0125] in, Let P represent the topographic residuals and surface deformation parameters, P1 represent the initial weight matrix of the initial solution of the topographic residuals and surface deformation parameters, and P1 represent the initial observation phase weight matrix of the first repeated orbit. i P represents the initial observation phase weight matrix for the i-th repeating orbit. n Let represent the initial observation phase weight matrix for the nth repeating orbit, i = 1, 2, ..., n, where n represents the total number of repeating orbits. T This represents the matrix transpose operation. Represents the parameter matrix, B represents the transformation matrix between parameters and unwrapped phase. geo Represents the projection transformation matrix. This represents the unwrapping phase matrix for all repeating orbits.

[0126] It should be noted that by projecting the radar line-of-sight deformation parameter vector corresponding to the repeating orbit onto the east-west and vertical directions, the elements in the terrain residual and surface deformation parameters in the phase equation are independent of the number of repeating orbits, thereby increasing the degree of freedom of the observation equation set.

[0127] For example, computer software such as Mathematica can be used to construct a set of observation equations and obtain the initial solutions of the topographic residuals and surface deformation parameters in the set of observation equations.

[0128] It is worth mentioning that the observation equations are constructed based on the unwrapped phase matrix of multiple repeating orbits, which increases the amount of data contained in the observation equations and thus improves the accuracy of the observation equations. At the same time, the observation equations are only expressions of the relationship between the unwrapped phase and various parameters, and cannot represent the terrain deformation. Therefore, the initial solutions of the terrain residual and surface deformation parameters are obtained, which represent the relationship between terrain deformation and other parameters, and facilitate subsequent calculations.

[0129] Step 14: Robustly estimate the initial weight matrix in the initial solution to obtain the robust weight matrix.

[0130] In some embodiments of this application, the step of robustly estimating the initial weight matrix in the initial solution to obtain the robust weight matrix is ​​specifically as follows:

[0131] The first step is through the formula:

[0132]

[0133] Calculate the observation residual vector matrix V for the i-th repeating orbit. i ;

[0134] Among them, V i =[V i,1 ,...,V i,r ,...,V i,R ], V i,1 V represents the observation residual vector of the first observation phase of the i-th repeating orbit. i,r V represents the observation residual vector of the r-th observation phase of the i-th repeating orbit. i,R Let r represent the observation residual vector of the R-th observation phase of the i-th repeating orbit, where r = 1, 2, ..., R, and R represents the total number of observation phases. This represents the parameter matrix of the i-th repeating orbit. This represents the initial solution for topographic residuals and surface deformation parameters. Let represent the unwrapping phase matrix of the i-th repeating orbit.

[0135] The second step is to use the formula:

[0136]

[0137]

[0138] Calculate the normalized observation residual vector matrix of the i-th repeating orbit.

[0139] Where ξ represents the robust estimation parameter, and median() represents the calculation of the median. This represents the standard observation residual of the first observation phase of the i-th repeating orbit. This represents the standard observation residual of the r-th observation phase of the i-th repeating orbit. It represents the standard observation residual of the R-th observation phase of the i-th repeating orbit.

[0140] The third step is to use the formula:

[0141]

[0142] Calculate the robust phase weight of the r-th observation phase of the i-th repeating orbit.

[0143] Among them, P i,r P represents the initial observation phase weight of the r-th observation phase of the i-th repeating orbit. i,r ∈P i Both k0 and k1 are constants, and k0 <k1;

[0144] Obtaining a robust weight matrix

[0145] in, This represents the robust observation phase weight matrix for the first repeating orbit. The robust observation phase weight represents the first observation phase of the first repeating orbit. This represents the robust observation phase weight for the r-th observation phase of the first repeating orbit. This represents the robust observation phase weight for the R-th observation phase of the first repeating orbit. This represents the robust observation phase weight matrix for the i-th repeating orbit. This represents the robust observation phase weight of the first observation phase of the i-th repeating orbit. This represents the robust observation phase weight for the R-th observation phase of the i-th repeating orbit. This represents the robust observation phase weight matrix for the nth repeating orbit. This represents the robust observation phase weight of the first observation phase of the nth repeating orbit. This represents the robust observation phase weight for the r-th observation phase of the n-th repeating orbit. This represents the robust observation phase weight of the Rth observation phase of the nth repeating orbit.

[0146] It should be noted that the elements in the initial weight matrix mentioned above can be coherent. The initial weight matrix includes the initial observation phase weight matrix for all repeating orbits, and the initial observation phase weight matrix for each repeating orbit includes the initial observation phase weights for all observation phases of that repeating orbit.

[0147] For example, robust estimation of the initial weight matrix can be performed using computer software such as MATLAB.

[0148] It is worth mentioning that robust estimation of the initial weight matrix can reduce the error of the elements in the initial weight matrix.

[0149] Step 15: Calculate the posterior variance matrix corresponding to the target region using the initial solution and the robust weight matrix, and update the robust weight matrix using the posterior variance matrix to obtain the weight matrix.

[0150] In some embodiments of this application, the steps described above—calculating the posterior variance matrix corresponding to the target region using the initial solution and the robust weight matrix, and updating the robust weight matrix using the posterior variance matrix to obtain the weight matrix—are specifically as follows:

[0151] The first step is to calculate the posterior variance matrix corresponding to the target region using the initial solution and the robust weight matrix.

[0152] Specifically, through the formula:

[0153]

[0154] Calculate the posterior variance matrix corresponding to the target region.

[0155] in, This represents the posterior variance of the observation residual vector for the first repeated orbit. This represents the posterior variance of the observation residual vector for the second repeated orbit. Let Vn represent the posterior variance of the observation residual vector of the nth repeating orbit, and V1 represent the observation residual vector matrix of the 1st repeating orbit. n Let s represent the observation residual vector matrix of the nth repeating orbit, j = 1, 2, ..., n. 1,1 s 1,n s i,1 s i,i s i,ns n,1 and s n,n All represent posterior variance parameters:

[0156] s i,i =n i -2tr(N -1 N i )+tr(N -1 N i N -1 N i )

[0157] s i,j =n i -tr(N -1 N i N -1 N j )

[0158]

[0159] N = N1 + N2 + ... + N n

[0160] Where, n i N represents i The i-th matrix element in the matrix, tr() represents the trace of the matrix, N represents the coefficient matrix of the normal equation, N i N represents the normal equation coefficient matrix corresponding to the i-th repeating orbit. j This represents the normal equation coefficient matrix corresponding to the j-th repeating orbit. Ni represents the coefficient matrix corresponding to the i-th repeating orbit, N1 represents the normal equation coefficient matrix corresponding to the 1st repeating orbit, N2 represents the normal equation coefficient matrix corresponding to the 2nd repeating orbit, and Ni represents the normal equation coefficient matrix corresponding to the i-th repeating orbit. n Let N = [N1, ..., N] represent the normal equation coefficient matrix corresponding to the nth repeating orbit. n ].

[0161] The second step is to update the robust weight matrix using the posterior variance matrix to obtain the weight matrix.

[0162] Specifically, through the formula:

[0163]

[0164] The elements of the initial weight matrix are updated to obtain the final weight matrix.

[0165] in, This represents the observation phase weight matrix for the first repeating orbit. This represents the observation phase weight matrix for the i-th repeating orbit. This represents the observation phase weight matrix for the nth repeating orbit.

[0166] For example, the above calculations can be performed using computer software for data computation such as Mathematica and MATLAB to obtain the weight matrix.

[0167] It is worth mentioning that by calculating the posterior variance matrix and updating the robust weight matrix, the weights between the observation residual vectors of each repeating orbit can be accurately estimated, while the impact of the unwrapping phase error on the accuracy of the weight calculation can be reduced.

[0168] Step 16: Determine whether the elements in the posterior variance matrix are equal. If so, use the weight matrix as the optimal weight matrix; otherwise, use the weight matrix as the initial weight matrix in the initial solution and return to the step of performing robust estimation on the initial weight matrix in the initial solution to obtain the robust weight matrix.

[0169] For example, if all elements in the posterior variance matrix are equal, then the weight matrix at this time is taken as the optimal weight matrix. If all elements in the posterior variance matrix are not equal, then the weight matrix at this time is taken as the initial weight matrix, and a robust estimation is performed on the initial weight matrix to calculate the posterior variance matrix. If all elements in the posterior variance matrix are equal at this time, it means that the standard weight matrix at this time has reached the expected accuracy and is the optimal weight matrix.

[0170] It should be noted that due to the influence of spatiotemporal decoherence and atmospheric disturbances, there are differences in the accuracy between unwrapped phase data. Therefore, it is unreasonable to set the values ​​of the elements in the weight matrix of the observation equation set to be equal. At the same time, the accuracy of the initial weight matrix is ​​low and cannot meet the requirements. Therefore, it is necessary to perform robust estimation and updating of the initial weights.

[0171] It is worth mentioning that taking the weight matrix in which all elements in the posterior variance matrix are equal as the optimal weight matrix can yield the optimal weight matrix that achieves the expected accuracy.

[0172] Step 17: Update the observation equations using the optimal weight matrix, and estimate the terrain residuals of the target area at time T using the updated observation equations.

[0173] The aforementioned terrain residuals are the terrain deformation of the target area over a period of time. The time interval described by the terrain residuals depends on the time interval of the SAR image data acquired from multiple repeating orbits.

[0174] In some embodiments of this application, the observation equation set can be updated by using the optimal weight matrix as the weight matrix in the observation equation set, and then the terrain residual can be estimated by using least squares to calculate the observation equation set.

[0175] Specifically, through the formula:

[0176]

[0177] The deformation parameter vector matrix is ​​calculated. and the terrain residual Δh at time T;

[0178] in, Represents the parameter matrix, B represents the transformation matrix between parameters and unwrapped phase. geo Let P represent the projection transformation matrix. * This represents the optimal weight matrix. Represents the unwrapped phase matrix of all repeating orbits, () T This represents the matrix transpose operation.

[0179] It should be noted that the result of the above formula is a vector matrix, and the first four rows of this vector matrix are the deformation parameter vector matrix. The element in the 5th row is the terrain residual Δh.

[0180] For example, the above steps can be performed using data processing computer software such as Matlab to estimate the terrain residuals.

[0181] It is worth mentioning that the terrain residual obtained by using the optimal weight matrix has low error and can accurately describe the terrain deformation of the target area over a period of time.

[0182] Step 18: Add the terrain residual of the target area at time T to the external reference DEM of the target area to obtain the DEM of the target area at the current time.

[0183] For example, if the available external reference DEM for the target area corresponds to a time one year ago, then by acquiring the SAR image of the target area at the most recent time, calculating the terrain residual at the most recent time, and finally adding the terrain residual to the external reference DEM from one year ago, the DEM of the target area at the current time can be obtained, thus realizing the updating of the DEM.

[0184] It should be noted that the DEM can be updated using computer software such as MATLAB. The DEM can be a TanDEM-X DEM acquired by the TerraSAR-X / TanDEM-X system.

[0185] It is worth mentioning that updating the external reference DEM of the target area using accurate terrain residuals results in a highly accurate DEM at the current moment.

[0186] The following example illustrates the high-frequency DEM update method based on repeating orbit InSAR of this application.

[0187] Using the Hambach open-pit coal mine in Germany as the experimental area, Sentinel-1 SAR datasets covering the mining area from March to June 2019 were selected, including two ascending orbits and two descending orbits. The ascending orbits had an incident angle of approximately 33.5 degrees and an azimuth of approximately -11.2 degrees (counterclockwise rotation), while the descending orbits had an incident angle of approximately 43.6 degrees and an azimuth of approximately -170.8 degrees (counterclockwise rotation). TanDEM-X with a spatial resolution of 30m was used as the external reference DEM (i.e., the DEM from the first time point mentioned above).

[0188] The accuracy of the DEM high-frequency update method based on repeating orbit InSAR provided in this application was verified by using GEDI satellite altimetry data covering the interior of the mine pit in the experimental area. The corresponding date of the GEDI data is April 20, 2019.

[0189] To demonstrate the advantages of the method provided in this application, its accuracy is compared with that of the traditional single-track DEM update method. The error histogram is shown below. Figure 2 As shown in the figure, the horizontal axis represents the elevation error (meters), which is the error between the updated DEM and the actual terrain data, with each grid representing 3 meters. The vertical axis represents the frequency, with each grid representing 0.05. The white histogram represents the error of the updated DEM obtained using the traditional single-track DEM update method, while the black histogram represents the error of the updated DEM obtained using the high-frequency DEM update method based on repeated-track InSAR of this invention, distributed within the error range of 0 to 20 meters. Its root mean square error (RMSE) has been improved from 21.8m to 8.7m, representing an accuracy improvement of approximately 60%.

[0190] Therefore, the high-frequency DEM update method based on repeating orbit InSAR proposed in this application can effectively improve the accuracy of terrain estimation, and can still improve the accuracy of DEM update when the terrain of the target area changes dynamically in a short time series.

[0191] The following is an exemplary description of the DEM high-frequency update device based on repeating orbit InSAR provided in this application.

[0192] like Figure 3 As shown, this application embodiment provides a DEM high-frequency update device based on repeating orbit InSAR. The DEM high-frequency update device 300 based on repeating orbit InSAR includes:

[0193] The acquisition module 301 acquires SAR image data of the target area at multiple repeating orbits at time T; the SAR image data includes SAR images of multiple observation phases, with time T being the current time.

[0194] The unwrapping module 302 is used to unwrap the SAR image data for each repeating orbit using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit.

[0195] Module 303 is used to construct the observation equation set based on the unwrapped phase matrix of all repeating orbits, and to obtain the initial solutions of the terrain residual and surface deformation parameters in the observation equation set.

[0196] Calculation module 304 is used to robustly estimate the initial weight matrix in the initial solution to obtain a robust weight matrix;

[0197] The update module 305 calculates the posterior variance matrix corresponding to the target region using the initial solution and the robust weight matrix, and updates the robust weight matrix using the posterior variance matrix to obtain the weight matrix.

[0198] Decision module 306 determines whether the elements in the posterior variance matrix are equal. If they are, the weight matrix is ​​used as the optimal weight matrix; otherwise, the weight matrix is ​​used as the initial weight matrix in the initial solution, and the step of robustly estimating the initial weight matrix in the initial solution to obtain the robust weight matrix is ​​returned.

[0199] The estimation module 307 updates the observation equation set using the optimal weight matrix and estimates the terrain residual of the target area at time T using the updated observation equation set.

[0200] The addition module 308 is used to add the terrain residual of the target area at time T to the external reference DEM of the target area to obtain the DEM of the target area at the current time.

[0201] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0202] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0203] like Figure 4 As shown, an embodiment of this application provides a terminal device, wherein the terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 4 The diagram shows only one processor, a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 executes the computer program D102 to implement the steps in any of the above method embodiments.

[0204] Specifically, when the processor D100 executes the computer program D102, it acquires SAR image data of multiple repeating orbits of the target area at T time points. Then, for each repeating orbit, it uses DInSAR technology to unwrap the SAR image data, obtaining the unwrapped phase matrix of the repeating orbit. Based on the unwrapped phase matrices of all repeating orbits, it constructs an observation equation set, which increases the amount of data contained in the observation equation set and thus improves its accuracy. Then, it obtains the initial solutions for the terrain residuals and surface deformation parameters in the observation equation set, performs robust estimation on the initial weight matrix in the initial solution to obtain a robust weight matrix, and uses the initial solution and robust weight matrix to calculate the posterior variance matrix corresponding to the target area. Finally, it uses the posterior variance matrix to update the robust weight matrix. The algorithm retrieves the weight matrix and then checks if the elements in the posterior variance matrix are equal. If they are, the weight matrix is ​​used as the optimal weight matrix; otherwise, it is used as the initial weight matrix in the initial solution. The algorithm then returns to the step of robustly estimating the initial weight matrix in the initial solution to obtain a robust weight matrix. By using robust estimation and the posterior variance matrix, the error of the optimal weight matrix can be reduced and its accuracy improved. The optimal weight matrix is ​​then used to update the observation equations, and the updated observation equations are used to estimate the terrain residual of the target area at time T. This significantly reduces the error of the terrain residual and improves accuracy. Finally, when the terrain residual of the target area at time T is added to the external reference DEM of the target area to obtain the DEM of the target area at the current time, the accuracy of the DEM update is effectively improved.

[0205] Furthermore, this application acquires SAR image data from multiple repeating orbits, and the SAR image data includes SAR images from multiple observation phases at multiple times. This decomposes the terrain deformation of the target area over a period of time into terrain at multiple times, which is suitable for DEM updates in short time series and ensures the amount of data in short time series, thus achieving accurate DEM updates in short time series.

[0206] The processor D100 can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0207] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may be an external storage device of the terminal device D10, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device D10. Furthermore, the memory D101 may include both internal and external storage units of the terminal device D10. The memory D101 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory D101 can also be used to temporarily store data that has been output or will be output.

[0208] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0209] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0210] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / terminal for a DEM high-frequency update method based on repeating orbit InSAR, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0211] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0212] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0213] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A high-frequency DEM update method based on repeating orbit InSAR, characterized in that, include: Obtain the target area SAR image data of multiple repeating orbits at a given time; the SAR image data includes SAR images of multiple observation phases, the first... The current moment is represented by the current moment. For each repeating orbit, the SAR image data is unwrapped using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit. The observation equations are constructed based on the unwrapped phase matrix of all repeating orbits, and the initial solutions of the topographic residuals and surface deformation parameters in the observation equations are obtained. A robust weight matrix is ​​obtained by robustly estimating the initial weight matrix in the initial solution; The posterior variance matrix corresponding to the target region is calculated using the initial solution and the robust weight matrix, and the robust weight matrix is ​​updated using the posterior variance matrix to obtain the weight matrix. Determine whether the elements in the posterior variance matrix are equal. If so, use the weight matrix as the optimal weight matrix; otherwise, use the weight matrix as the initial weight matrix in the initial solution and return to the step of robustly estimating the initial weight matrix in the initial solution to obtain the robust weight matrix. The observation equations are updated using the optimal weight matrix, and the updated observation equations are used to estimate the target region in the [missing information]. Terrain residuals at each moment; The target area is in the first The terrain residual at each time step is added to the external reference DEM of the target area to obtain the DEM of the target area at the current time step.

2. The DEM high-frequency update method according to claim 1, characterized in that, The initial solutions for the topographic residuals and surface deformation parameters are: in, This represents the topographic residual and surface deformation parameters. This represents the initial weight matrix of the initial solution for the terrain residuals and surface deformation parameters. This represents the initial observation phase weight matrix for the first repeating orbit. Indicates the first The initial observation phase weight matrix for each repeating orbit. Indicates the first The initial observation phase weight matrix for each repeating orbit. , This represents the total number of repeating orbitals. This represents the matrix transpose operation. Represents the parameter matrix, , This represents the transformation matrix between the parameters and the unwrapped phase. Represents the projection transformation matrix. This represents the unwrapping phase matrix for all repeating orbits.

3. The DEM high-frequency update method according to claim 2, characterized in that, The robust weight matrix is ​​obtained by robustly estimating the initial weight matrix in the initial solution, including: Through the formula: Calculate the first The observation residual vector matrix of each repeating orbit ; in, , Indicates the first The observation residual vector of the first observation phase of a repeating orbit. Indicates the first The first repeating orbit The observation residual vector for each observation phase, Indicates the first The first repeating orbit The observation residual vector for each observation phase, , This represents the total number of observed phases. Indicates the first The parameter matrix of repeating orbits, This represents the initial solution for the topographic residuals and surface deformation parameters. Indicates the first The unwrapping phase matrix of a repeating orbit; Through the formula: Calculate the first The normalized observation residual vector matrix of repeated orbits ; in, Indicates robust estimation parameters, This indicates the calculation of the median. , Indicates the first The standard observation residual of the first observation phase of a repeating orbit Indicates the first The first repeating orbit The standard observation residuals for each observation phase Indicates the first The first repeating orbit Standard observation residuals for each observation phase; Through the formula: Calculate the first The first repeating orbit Robust phase weights for each observation phase ; in, Indicates the first The first repeating orbit The initial observation phase weights for each observation phase. , and All are constants, and ; Obtain the robust weight matrix ; in, , This represents the robust observation phase weight matrix of the first repeating orbit. This represents the robust observation phase weight of the first observation phase of the first repeating orbit. The first repeating orbit represents the first... Robust observation phase weights for each observation phase The first repeating orbit represents the first... Robust observation phase weights for each observation phase , Indicates the first Robust observation phase weight matrix for repeating orbits, Indicates the first Robust observation phase weights for the first observation phase of a repeating orbit. Indicates the first The first repeating orbit Robust observation phase weights for each observation phase , Indicates the first Robust observation phase weight matrix for repeating orbits, Indicates the first Robust observation phase weights for the first observation phase of a repeating orbit. Indicates the first The first repeating orbit Robust observation phase weights for each observation phase Indicates the first The first repeating orbit Robust observation phase weights for each observation phase.

4. The DEM high-frequency update method according to claim 3, characterized in that, The step of calculating the posterior variance matrix corresponding to the target region using the initial solution and the robust weight matrix includes: Through the formula: Calculate the posterior variance matrix corresponding to the target region. ; in, This represents the posterior variance of the observation residual vector of the first repeated orbit. This represents the posterior variance of the observation residual vector for the second repeated orbit. Indicates the first The posterior variance of the observation residual vector of a repeating orbit This represents the observation residual vector matrix of the first repeated orbit. Indicates the first The observation residual vector matrix of repeating orbits, , , , , , , and All represent posterior variance parameters: in, express The first in matrix elements, Represents finding the trace of a matrix. Representing the coefficient matrix of the normal equation, Indicates the first The normal equation coefficient matrix corresponding to each repeating orbit. Indicates the first The normal equation coefficient matrix corresponding to each repeating orbit. Indicates the first The coefficient matrix corresponding to each repeating orbital This represents the normal equation coefficient matrix corresponding to the first repeating trajectory. The normal equation coefficient matrix corresponding to the second repeating orbit Indicates the first The normal equation coefficient matrix corresponding to each repeating orbit. .

5. The DEM high-frequency update method according to claim 4, characterized in that, The step of updating the robust weight matrix using the posterior variance matrix to obtain the weight matrix includes: Through the formula: The elements in the initial weight matrix are updated to obtain the weight matrix. ; in, , This represents the observation phase weight matrix for the first repeating orbit. Indicates the first The observed phase weight matrix of each repeating orbit. Indicates the first The observation phase weight matrix of each repeating orbit.

6. The DEM high-frequency update method according to claim 1, characterized in that, The updated observation equations are used to estimate the target region in the [number]th [period]. The terrain residuals at each time point include: Through the formula: The deformation parameter vector matrix is ​​calculated. and the first Terrain residuals at each moment ; in, Represents the parameter matrix, , This represents the transformation matrix between the parameters and the unwrapped phase. Represents the projection transformation matrix. This represents the optimal weight matrix. This represents the unwrapped phase matrix for all repeating orbits. This represents the matrix transpose operation.

7. The DEM high-frequency update method according to claim 1, characterized in that, The step of unwrapping the SAR image data using DInSAR technology to obtain the unwrapped phase matrix of the repeating orbit includes: Multiple interferometric pairs in the SAR image data are generated using DInSAR technology; Using the external reference DEM of the target region, differential processing is performed on each interference pair to obtain differential interference pairs; The differential interference pair is unwrapped to obtain the unwrapped phase, and the unwrapped phase matrix of the repeating orbit is obtained.

8. The DEM high-frequency update method according to claim 6, characterized in that, The observation equations constructed based on the unwrapped phase matrix of all repeating orbits include: Based on the unwrapped phase matrix of all repeating orbits, an observation equation is constructed between the phase and the aforementioned topographic residuals and surface deformation parameters: in, This represents the coefficient matrix of the first repeating orbit. Indicates the first The coefficient matrix of repeating orbits, , and Both represent the matrix dimension. express The total number of differential interference pairs of repeating orbits. Represents the deformation parameter vector matrix, , This represents the deformation parameter vector of the first repeating orbit. This represents the deformation parameter vector of the second repeating orbit. Indicates the first The deformation parameter vector of a repeating orbit. This represents the matrix transpose operation. Indicates the residual phase. , This represents the unwrapped phase matrix of the first repeating orbit. This represents the unwrapping phase of the first differential interference pair of the first repeating orbit. This represents the unwrapping phase of the second differential interference pair of the first repeating orbit. The first repeating orbit represents the first... The unwrapping phase of a differential interference pair , Indicates the first The unwrapping phase matrix of repeating orbits Indicates the first The unwrapping phase of the first differential interference pair of repeating orbits Indicates the first The unwrapping phase of the second differential interference pair of repeating orbits. Indicates the first The first repeating orbit The unwrapping phase of a differential interference pair; For each of the multiple repeating orbits, the radar line-of-sight deformation parameter vector corresponding to the repeating orbit is projected onto the east-west and vertical directions, resulting in the observation equation set as follows: in, , and Both represent the matrix dimension. Represents topographic residuals and surface deformation parameters. , This represents the average rate of deformation in the vertical direction. This represents the average rate of deformation along the east-west direction. This represents the vertical deformation acceleration. Indicates the deformation acceleration in the east-west direction. This represents the unwrapped phase matrix for all repeating orbits. This represents the transformation matrix between the parameters and the unwrapped phase. Representing the projection transformation matrix: in, This represents the angle of incidence of the satellite on the first repeating orbit. This indicates the angle of incidence of the satellite on the second repeating orbit. This indicates the flight azimuth angle of the first repeating orbit satellite. This indicates the flight azimuth angle of the second repeating orbit satellite. This represents the projection relationship between the slant range direction of the satellite and the vertical direction of the Earth's surface on the first repeating orbit. This indicates the projection relationship between the slant range direction of the satellite and the vertical direction of the Earth's surface on the second repeating orbit. This represents the projection relationship between the slant range direction of the satellite on the first repeating orbit and the east-west direction of the Earth's surface. This indicates the projection relationship between the slant range of the satellite on the second repeating orbit and the east-west direction of the Earth's surface.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the high-frequency DEM update method based on repeating orbit InSAR as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the high-frequency DEM update method based on repeating orbit InSAR as described in any one of claims 1 to 8.

Citation Information

Patent Citations

  • Synthetic aperture radar image orientation secondary focusing method

    CN104237885A

  • InSAR residual motion error estimation method and device considering topographic relief

    CN111273293A