InSAR landslide deformation time series monitoring method and storage medium integrating CR and GNSS

By integrating CR and GNSS technology, the accuracy problem of large-deformed landslide deformation monitoring in InSAR technology is solved, the true reflection of landslide deformation characteristics and evolution process analysis are realized, and the measurement ability of the maximum deformation gradient is improved.

CN117761716BActive Publication Date: 2024-11-22CHINA AERO GEOPHYSICAL SURVEY & REMOTE SENSING CENT FOR LAND & RESOURCES
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311775084.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-21
Publication Date
2024-11-22
Estimated Expiration
2043-12-21

AI Technical Summary

Technical Problem

The existing InSAR technology is difficult to accurately monitor the deformation evolution process of large deformation landslides, resulting in blurred phase of the interference pattern and the inability to restore the deformation phase.

Method used

Fusion of CR and GNSS technology, convert geographical coordinates into SAR distance Doppler coordinates through GNSS devices, identify CR positions, and convert GNSS observations into radar line-of-sight observations, determine the time baseline threshold, and realize seamlessness of differential interference maps. Connection and deformation timing analysis.

Benefits of technology

The measurement ability of the maximum deformation gradient is improved, and the maximum deformation level of the landslide is increased from tens of centimeters to several meters, realizing the true reflection of the deformation characteristics of large-deformed landslides and analysis of the evolution process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117761716B_ABST
    Figure CN117761716B_ABST
Patent Text Reader

Abstract

The present invention discloses an InSAR landslide deformation time series monitoring method and storage medium integrating CR and GNSS. According to an embodiment of the present invention, at least one set of CR and GNSS equipment is arranged at the same position of the landslide, and the geographic coordinates recorded by the GNSS equipment are converted into SAR range Doppler coordinates, thereby identifying the position of the CR on the SAR main image, and then the GNSS equipment observation value is converted into the radar line of sight observation value and the time baseline threshold is determined, and then the differential interference graph that can both meet the coherence and restore the correct deformation information is determined according to the time baseline threshold, and the interference point target analysis is performed based on multiple differential interference graphs to obtain the deformation rate and deformation time series of the coherent point target. Therefore, the present invention can seamlessly connect CR and GNSS positioning, deformation time series observation on GNSS points, and high-resolution InSAR extraction of surface deformation information, so as to realize the deformation evolution process analysis of large deformation landslides.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster monitoring, and more specifically, to an InSAR landslide deformation time series monitoring method and storage medium integrating CR and GNSS. Background Art

[0002] The principle of InSAR deformation monitoring is to use the differential interference processing of two satellite-borne SAR images before and after the change to obtain the entangled phase. After phase unwrapping, the absolute phase is obtained to characterize the surface elevation change information. If a large deformation occurs in a small area, it often causes a sharp change in the phase in the interference pattern, making the interference fringes blurred, and thus the deformation phase cannot be accurately restored. In other words, the existing technology lacks a method for monitoring and analyzing the deformation evolution process of large deformation landslides. Summary of the invention

[0003] The embodiments of the present invention provide an InSAR landslide deformation time-series monitoring method and storage medium that integrates CR and GNSS, seamlessly connects CR and GNSS positioning, deformation time-series observation on GNSS points, and high-resolution InSAR extraction of surface deformation information, thereby improving the maximum deformation gradient capability that can be detected, and increasing the maximum deformation magnitude of the measured landslide from tens of centimeters to several meters, which can more realistically reflect the deformation characteristics of the landslide and realize the analysis of the deformation evolution process of large-deformation landslides.

[0004] According to a first aspect of the present invention, an embodiment of the present invention provides an InSAR landslide deformation time series monitoring method integrating CR and GNSS, the method comprising: obtaining satellite radar data covering a landslide area, and selecting a SAR main image from the satellite radar data; obtaining data sent from at least one set of CR and GNSS equipment deployed in an area with significant landslide deformation, wherein at least one set of CR and GNSS equipment is deployed at the same position of the landslide for accuracy verification; converting the geographic coordinates recorded by the GNSS equipment into range Doppler coordinates of the SAR main image, and identifying the position of the CR on the SAR main image; calculating satellite orbit information covering the landslide area according to a parameter file of the SAR main image, the satellite orbit information including a satellite orbit azimuth and a radar incident angle; converting GNSS equipment observation values ​​into radar line-of-sight observation values ​​according to the satellite orbit information; determining a radar line-of-sight deformation time series based on the radar line-of-sight observation values, and calculating a radar line-of-sight deformation time series based on the radar line-of-sight deformation time series. Calculate the theoretical radar satellite shooting time interval and set the time baseline threshold; perform sub-pixel registration on all images in the satellite radar data to the range Doppler coordinate system of the main image to obtain the registered SAR image; select images with a time interval not exceeding the time baseline threshold from the registered SAR image as interference image pairs, and perform differential interference processing on the interference image pairs to obtain a differential interference atlas; calculate the initial line of sight deformation rate according to the differential interference atlas; for adjacent images with a time interval exceeding the time baseline threshold in the registered SAR image, determine the alternative differential interference map of the adjacent images according to the initial line of sight deformation rate and the time interval of the adjacent images; use the differential interference atlas and the alternative differential interference map as standard data input of the interference point target analysis method IPTA, select coherent point targets including CR points, perform deformation time series analysis on the differential interference atlas and the alternative differential interference map, and obtain the deformation rate and deformation time series of the coherent point target.

[0005] The above-mentioned embodiment of the present invention can identify the position of CR on the SAR main image by placing at least one set of CR and GNSS equipment at the same position of the landslide, and converting the geographical coordinates recorded by the GNSS equipment into SAR range Doppler coordinates, and then converting the GNSS equipment observation value into the radar line of sight observation value and determining the time baseline threshold, and then determining the differential interference graph that can both meet the coherence and restore the correct deformation information according to the time baseline threshold, and performing interference point target analysis based on multiple differential interference graphs to obtain the deformation rate and deformation time series of the coherent point target. Thus, the CR and GNSS positioning, the deformation time series observation on the GNSS point, and the high-resolution InSAR extraction of the deformation information on the surface are seamlessly connected, which can truly reflect the deformation characteristics of the landslide and realize the analysis of the deformation evolution process of the large deformation landslide.

[0006] In some embodiments of the present invention, the satellite radar data are different mode data covering the same landslide area, and the incident angles of the different mode data are substantially consistent.

[0007] In some embodiments of the present invention, according to the distance equation R=|R s -R t |, and the Doppler equation The geographic coordinates recorded by the GNSS device are converted into the range Doppler coordinates of the SAR main image; wherein R is the distance from the sensor to the point target, R s is the position of the sensor, v s is the velocity vector of the sensor, R t is the position of the point target, v t is the velocity vector of the point target, λ is the radar wavelength, f d is the Doppler frequency.

[0008] In some embodiments of the present invention, the GNSS device observation value is converted into the radar line of sight observation value according to the following calculation formula:

[0009] [Unsinα-Uecosα]sinθ+Uucosθ+δ los =g los

[0010] Among them, Ue, Un, and Uu are the east-west, north-south, and vertical deformation components of the GNSS equipment, α is the satellite orbit azimuth, θ is the radar incident angle, and g is the los The radar line of sight observation value is obtained by converting the GNSS equipment observation value.

[0011] In some embodiments of the present invention, calculating the theoretical radar satellite shooting time interval according to the radar line of sight deformation time series includes:

[0012] The radar line of sight deformation rate is calculated according to the following formula:

[0013]

[0014] Among them, t0 is the starting observation time, t i is the observation cut-off time, Δv glos From the start observation time t0 to the end observation time t i The radar line of sight deformation rate between them;

[0015] Δv glos Substitute Δv into the following calculation formula to calculate the theoretical radar satellite shooting time interval:

[0016]

[0017] Among them, Δv is the change value of the radar line of sight deformation rate in the same pixel, λ is the radar wavelength, and Δt is the theoretical radar satellite shooting time interval.

[0018] In some embodiments of the present invention, the initial sight line deformation rate is calculated according to the following deformation rate calculation formula:

[0019]

[0020] Among them, Δt k is the time interval of the kth interference image pair in the differential interference atlas, φ k is the unwrapping phase of the kth group of interference image pairs in the differential interference atlas, N is the total value of the interference image pairs, and p is the initial line-of-sight deformation rate.

[0021] In some embodiments of the present invention, determining an alternative differential interference pattern of adjacent images based on the initial line-of-sight deformation rate and the time interval between adjacent images includes: the product of the initial line-of-sight deformation rate and the time interval between adjacent images is the alternative line-of-sight deformation amount of the adjacent images; and determining an alternative differential interference pattern of the adjacent images based on the alternative line-of-sight deformation amount.

[0022] In some embodiments of the present invention, the method further comprises: filtering the differential interferogram set and the alternative differential interferogram using the Goldstein-Werner filtering method; performing spatial phase unwrapping on the filtered differential interferogram using the minimum cost flow algorithm; comprehensively selecting the coherent point target with stable SAR image spectrum characteristics using the amplitude dispersion index and the coherence coefficient threshold; and calculating the deformation rate and deformation time series of all coherent point targets according to the following formula:

[0023]

[0024] in, is the phase difference between the coherent point targets i and j in the kth interference image pair, θ is the incident angle, is the vertical baseline, t k is the time interval, is the relative elevation difference between neighboring targets i and j, is the relative deformation rate between neighboring targets i and j, is other phase, λ is radar wavelength, and R is the distance from sensor to point target. According to the deformation rate and deformation time series of the coherent point target, the interference point target analysis method IPTA is used to perform deformation time series analysis.

[0025] In some embodiments of the present invention, the method further comprises: calculating the root mean square error σ of the difference between the InSAR measurement value at the CR point and the line-of-sight deformation observation value converted by GNSS in the same time period according to the following formula:

[0026]

[0027] Among them, x i is the InSAR measurement value at point CR, y i is the deformation observation value converted from GNSS to line of sight at the CR point, and n is the number of statistical samples; the measurement accuracy of the InSAR result is judged by the root mean square error of the difference between the InSAR measurement value at the CR point and the deformation observation value converted from GNSS to line of sight.

[0028] According to the second aspect of the present invention, an embodiment of the present invention provides a computer-readable storage medium having computer-readable instructions stored thereon. When the computer-readable instructions are executed by a processor, the computer performs the following operations: the operations include the steps included in the method described in any one of the above embodiments.

[0029] In summary, the InSAR landslide deformation time-series monitoring method and storage medium integrating CR and GNSS provided in the embodiments of the present invention seamlessly connect CR and GNSS positioning, deformation time-series observation on GNSS points, and deformation information on the surface extracted by high-resolution InSAR, and can realize the deformation evolution process analysis of large deformation landslides. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flow chart of a method for time-series monitoring of landslide deformation by InSAR integrating CR and GNSS according to an embodiment of the present invention;

[0031] Figure 2 Is adopted Figure 1 The flowchart of a specific example of the InSAR landslide deformation time series monitoring method integrating CR and GNSS is shown;

[0032] Figure 3 It is the CR point map identified on the radar image;

[0033] Figure 4 It is the spatial and temporal baseline distribution diagram of 35 interferometric image pairs formed by 24 SAR images. The horizontal axis is time (year / month / day), the vertical axis is the vertical baseline, Bperp min and max are the minimum and maximum values ​​of the vertical baseline, and Delta_T min and max are the minimum and maximum values ​​of the time baseline;

[0034] Figure 5aIt is the filtered differential interferogram from 2023 / 04 / 04 to 2023 / 04 / 08 (4-day interval), where -π to π refers to a phase change of a 2π cycle;

[0035] Figure 5b It is the filtered differential interferogram from 2023 / 04 / 04 to 2023 / 04 / 12 (8-day interval), where -π to π refers to a phase change of a 2π cycle;

[0036] Figure 5c It is the filtered differential interferogram from 2023 / 04 / 04 to 2023 / 04 / 16 (12-day interval), where -π to π refers to a phase change of a 2π cycle;

[0037] Figure 6 It is the LT-1 satellite multi-mode InSAR interactive differential interferogram (2023 / 06 / 19_2023 / 06 / 27), that is, the 3m (2023 / 06 / 19) and 6m (2023 / 06 / 27) resolution SAR images are processed by differential interferometry;

[0038] Figure 7 is the InSAR deformation rate map of the Xieliupao landslide;

[0039] Figure 8a is the InSAR deformation time series of the typical coherent point target at the rear edge of the selected landslide in Xieliupao. InSAR_ori and InSAR_cor are the deformation time series obtained by calculating the original differential interferogram and the substituted differential interferogram of the adjacent images for more than 8 days, respectively.

[0040] Figure 8b It is the InSAR deformation time series of the typical coherent point target in the middle of the selected Xieliupao landslide. InSAR_ori and InSAR_cor are the deformation time series obtained by calculating the original differential interferogram and the substituted differential interferogram of the adjacent images for more than 8 days.

[0041] Figure 8c It is the InSAR deformation time series of the typical coherent point target at the front edge of the selected landslide of the leak slope. InSAR_ori and InSAR_cor are the deformation time series obtained by calculating the original differential interferogram and the substituted differential interferogram of the adjacent images for more than 8 days.

[0042] Fig. 9 This is a deformation time series diagram comparing the InSAR measurement value at point CR-2 and the GNSS converted into line-of-sight deformation observation value. DETAILED DESCRIPTION

[0043] The following is a detailed description of various aspects of the present invention in conjunction with the accompanying drawings and specific embodiments. It should be understood by those skilled in the art that the various embodiments described below are only used for illustration and are not intended to limit the scope of protection of the present invention. Various different combinations of the various embodiments of the present invention may also be performed without departing from the essence of the present invention.

[0044] The following is a brief description of the terms used in this article.

[0045] CR: Corner Reflector, corner reflector.

[0046] GNSS: Global Navigation Satellite System.

[0047] InSAR: Interferometric Synthetic Aperture Radar, synthetic aperture radar interferometry, refers to the interferometry technology using synthetic aperture radar.

[0048] SAR: Synthetic Aperture Radar.

[0049] IPTA: Interferometric Point Target Analysis, interference point target analysis method.

[0050] Perpendicular Baseline: Vertical baseline.

[0051] SLC: Single Look Complex, single-look complex image, belongs to Level-1 image.

[0052] Figure 1 It is a flow chart of a method for time-series monitoring of landslide deformation by InSAR integrating CR and GNSS according to an embodiment of the present invention.

[0053] like Figure 1 As shown, in one embodiment of the present invention, the landslide deformation time series monitoring method may include: step S101, step S102, step S103, step S104, step S105, step S106, step S107, step S108, step S109, step S110 and step S111, and the above steps are described in detail below.

[0054] In step S101, satellite radar data covering the landslide area is obtained, and a SAR main image is selected from the satellite radar data. In some embodiments, the satellite radar data is different mode data covering the same landslide area, and the incident angles of the different mode data are substantially the same.

[0055] In this embodiment, the satellite radar data is LT-1 satellite radar data. Further, mode 1 data in the LT-1 satellite radar data is selected as the main SAR image, and the interactive interference of different mode SAR images of the LT-1 satellite constellation is used for the first time to monitor the deformation of large deformation landslides. Among them, the Land Exploration-1 (LT-1) 01 group of satellites consists of 2 identical L-band fully polarized radar satellites. The satellite is designed with 6 observation modes, with 6m / 100km (mode 2) and 3m / 50km (mode 1) as the main modes of deformation monitoring. The theoretical strict regression period is 8 days for a single star and 4 days for a double star. The actual revisit period depends on the shooting plan and the selected mode, and has the single / double star re-orbit interference and double star single flyby interference capabilities. In addition, the LT-1 satellite is the L-band radar satellite constellation with the shortest revisit period so far, which is suitable for monitoring large deformation landslides.

[0056] In step S102, data sent from at least one set of CR and GNSS equipment deployed in the area with significant landslide deformation is obtained, wherein at least one set of CR and GNSS equipment is deployed at the same position of the landslide for accuracy verification. In some embodiments, the data sent by the GNSS equipment includes the geographic coordinates recorded by the GNSS equipment and the east-west Ue, north-south Un, and vertical Uu deformation components; CR and GNSS equipment are deployed in the area with significant (or maximum) landslide deformation according to prior knowledge.

[0057] In step S103, the geographical coordinates recorded by the GNSS device are converted into range Doppler coordinates of the SAR main image, and the position of the CR on the SAR main image is identified.

[0058] In some embodiments, the geographic coordinates recorded by the GNSS device are converted into the range Doppler coordinates of the SAR main image according to the following formula (1):

[0059] Distance equation: R = |R s -R t |

[0060] Doppler equation:

[0061] Where R is the distance from the sensor to the point target, R s is the position of the sensor, v s is the velocity vector of the sensor, R tis the position of the point target, v t is the velocity vector of the point target, λ is the radar wavelength, f d is the Doppler frequency.

[0062] At least one set of CR and GNSS equipment is deployed at the same location of the landslide. The geographic coordinates recorded by the GNSS equipment are converted into SAR range Doppler coordinates to preliminarily identify the position of the CR on the SAR main image. Then, the position of the CR row and column numbers is preliminarily identified on the SAR main image. The bright white square point on the SAR main image is confirmed to be the CR point through visual interpretation.

[0063] In step S104, satellite orbit information covering the landslide area is calculated according to the parameter file of the SAR main image, and the satellite orbit information includes a satellite orbit azimuth and a radar incident angle.

[0064] In step S105, the GNSS device observation value is converted into the radar line of sight observation value according to the satellite orbit information. In some embodiments, the GNSS device observation value is converted into the radar line of sight observation value according to the following calculation formula (2):

[0065] [Unsinα-Uecosα]sinθ+Uucosθ+δ los =g los (2)

[0066] Among them, Ue, Un, and Uu are the east-west, north-south, and vertical deformation components of the GNSS equipment, α is the satellite orbit azimuth, θ is the radar incident angle, and g is the vertical deformation component of the GNSS equipment, respectively. los It is the radar line of sight observation value converted from the GNSS equipment observation value.

[0067] In step S106, a radar line of sight deformation time series is determined based on the GNSS radar line of sight observation value, a theoretical radar satellite shooting time interval is calculated according to the radar line of sight deformation time series, and a time baseline threshold is set.

[0068] In some implementations, calculating the theoretical radar satellite shooting time interval according to the radar line of sight deformation time series comprises the following steps:

[0069] First, the radar line of sight deformation rate is calculated according to the following calculation formula (3):

[0070]

[0071] Among them, t0 is the starting observation time, t i is the observation cut-off time, Δv glos From the start observation time t0 to the end observation time t i The radar line of sight deformation rate between them;

[0072] Then, the Δv obtained by calculation formula (3) is glos Substitute Δv into the following calculation formula (4) to calculate the theoretical radar satellite shooting time interval:

[0073]

[0074] Among them, Δv is the change value of the deformation rate in the radar line of sight in the same pixel, λ is the radar wavelength, and Δt is the theoretical radar satellite shooting time interval. According to the above formula (4), the shorter the revisit period of the radar satellite, the more conducive it is to monitor large deformation phenomena.

[0075] Furthermore, the time baseline threshold is set to m days according to the theoretical radar satellite shooting time interval and the SAR data acquisition situation.

[0076] In step S107, all images in the satellite radar data are sub-pixel registered to the range Doppler coordinate system of the main image to obtain a registered SAR image.

[0077] In step S108, images whose time interval does not exceed the time baseline threshold are selected from the registered SAR images as interference image pairs, and the interference image pairs are subjected to differential interference processing to obtain differential interference atlases. The time baseline threshold determined by the theoretical radar satellite shooting time interval obtained in combination with formula (4) is used to determine the differential interference atlas that can both meet the coherence and restore the correct deformation information, and then the initial deformation rate is estimated based on the data in the differential interference atlas.

[0078] In step S109, the initial line-of-sight deformation rate is calculated according to the differential interference pattern set. In some embodiments, the initial line-of-sight deformation rate is calculated according to the following deformation rate calculation formula (5):

[0079]

[0080] Among them, Δt k is the time interval of the kth interference image pair in the interference image pair set, φ k is the unwrapping phase of the kth interference image pair in the interference image pair set, N is the total value of the interference image pairs, and p is the initial line-of-sight deformation rate.

[0081] In step S110, for the adjacent images whose time intervals in the registered SAR images exceed the time baseline threshold, an alternative differential interferogram of the adjacent images is determined according to the initial sight line deformation rate and the time interval of the adjacent images. In some embodiments, determining the alternative differential interferogram of the adjacent images according to the initial sight line deformation rate and the time interval of the adjacent images includes: the product of the initial sight line deformation rate and the time interval of the adjacent images is the alternative sight line deformation amount of the adjacent image; and determining the alternative differential interferogram of the adjacent image according to the alternative sight line deformation amount.

[0082] In step S111, the differential interference atlas set and the alternative differential interference atlas are used as standard data input for the interference point target analysis method IPTA, and coherent point targets including CR points are selected to perform deformation timing analysis on the differential interference atlas set and the alternative differential interference atlas to obtain the deformation rate and deformation time series of the coherent point target.

[0083] In some implementations, the deformation rate and deformation time series of the coherent point target are obtained according to the following steps:

[0084] Step (1), filtering the differential interference pattern set and the alternative differential interference pattern using a Goldstein-Werner filtering method;

[0085] Step (2), using a minimum cost flow algorithm to perform spatial phase unwrapping on the filtered differential interferogram;

[0086] Step (3), using the amplitude dispersion index and the coherence coefficient threshold to comprehensively select the coherent point target with stable SAR image spectrum characteristics;

[0087] Step (4), calculate the deformation rate and deformation time series of all coherent point targets according to the following formula (6):

[0088]

[0089] in, is the phase difference between the coherent point targets i and j in the kth interferometric image pair, θ is the radar incident angle, is the vertical baseline, t k is the time interval, is the relative elevation difference between neighboring targets i and j, is the relative deformation rate between neighboring point targets i and j, λ is the radar wavelength, R is the distance from the sensor to the point target, is other phase (residual phase), other phases include, for example, orbit error phase, atmospheric delay phase, noise phase, etc.;

[0090] Step (5), using the interference point target analysis method IPTA to perform deformation timing analysis on the deformation rate and deformation time series of the coherent point target.

[0091] In this embodiment, there is no fixed order between steps S101 and S102.

[0092] By adopting the above method of the embodiment of the present invention, by deploying at least one set of CR and GNSS equipment at the same position of the landslide, and converting the geographical coordinates recorded by the GNSS equipment into SAR range Doppler coordinates, the position of the CR on the SAR main image can be identified, and then the GNSS equipment observation value is converted into the radar line of sight observation value and the time baseline threshold is determined, and then the differential interference graph that can satisfy the coherence and avoid phase ambiguity to restore the correct deformation information is determined according to the time baseline threshold, and the interference point target analysis is performed based on multiple differential interference graphs to obtain the deformation rate and deformation time series of the coherent point target. Thus, the CR and GNSS positioning, the deformation time series observation on the GNSS point, and the high-resolution InSAR extraction of the deformation information on the surface are seamlessly connected, which improves the maximum deformation gradient capability that can be detected, and the maximum deformation magnitude of the measured landslide is increased from the original tens of cm to several meters, which can more realistically reflect the deformation characteristics of the landslide and realize the deformation evolution process analysis of the large deformation landslide.

[0093] In a further embodiment, the landslide deformation time series monitoring method further includes: calculating the root mean square error σ of the difference between the InSAR measurement value at the CR point and the GNSS converted line-of-sight deformation observation value in the same time period according to the following formula (7):

[0094]

[0095] Among them, x i is the InSAR measurement value at point CR, y i is the deformation observation value converted from GNSS to line of sight at the CR point, and n is the number of statistical samples; the measurement accuracy of the InSAR result is judged by the root mean square error of the difference between the InSAR measurement value at the CR point and the deformation observation value converted from GNSS to line of sight.

[0096] The present invention provides a method of using Figure 1 The InSAR landslide deformation time series monitoring method integrating CR and GNSS is shown in the figure. It is a specific example of realizing the time series monitoring of large deformation landslide. This example takes the monitoring of Xieliupo landslide in Zhouqu, Gansu as an example. It uses LT-1 satellite mode 1 and mode 2 data and CR and GNSS equipment deployed on the landslide body for joint observation. Figures 2 to 9 To explain in detail:

[0097] Figure 2 Is adopted Figure 1The flowchart of a specific example of the InSAR landslide deformation time series monitoring method integrating CR and GNSS is shown; Figure 3 It is the CR point map identified on the radar image; Figure 4 It is the spatial and temporal baseline distribution diagram of 35 interferometric image pairs formed by 24 SAR images. The horizontal axis is time (year / month / day), the vertical axis is the vertical baseline, Bperp min and max are the minimum and maximum values ​​of the vertical baseline, and Delta_T min and max are the minimum and maximum values ​​of the time baseline; Figure 5a It is the filtered differential interferogram from 2023 / 04 / 04 to 2023 / 04 / 08 (4-day interval); Figure 5b It is the filtered differential interferogram from 2023 / 04 / 04 to 2023 / 04 / 12 (8-day interval); Figure 5c It is the filtered differential interferogram from 2023 / 04 / 04 to 2023 / 04 / 16 (12-day interval); Figure 6 It is the LT-1 satellite multi-mode InSAR interactive differential interferogram (2023 / 06 / 19_2023 / 06 / 27), that is, the 3m (2023 / 06 / 19) and 6m (2023 / 06 / 27) resolution SAR images are processed by differential interferometry; Figure 7 is the InSAR deformation rate map of the Xieliupao landslide; Figure 8a is the InSAR deformation time series of the typical coherent point target at the rear edge of the selected landslide in Xieliupao. InSAR_ori and InSAR_cor are the deformation time series obtained by calculating the original differential interferogram and the substituted differential interferogram of the adjacent images for more than 8 days, respectively. Figure 8b It is the InSAR deformation time series of the typical coherent point target in the middle of the selected Xieliupao landslide. InSAR_ori and InSAR_cor are the deformation time series obtained by calculating the original differential interferogram and the substituted differential interferogram of the adjacent images for more than 8 days. Figure 8c It is the InSAR deformation time series of the typical coherent point target at the front edge of the selected landslide of the leak slope. InSAR_ori and InSAR_cor are the deformation time series obtained by calculating the original differential interferogram and the substituted differential interferogram of the adjacent images for more than 8 days. Fig. 9 This is a deformation time series diagram comparing InSAR results and GNSS observations.

[0098] like Figure 2 As shown, the method includes:

[0099] Step 1: LT-1 satellite single-view complex image selection

[0100] The LT-1 satellite radar data covering the Xieliupo landslide were selected, including 6 scenes of mode 1 data (June 27, 2023-August 18, 2023) and 18 scenes of mode 2 data (March 15, 2023-June 19, 2023). The main parameters of the radar data are shown in Table 1 below.

[0101] Table 1

[0102] Radar parameters Mode 1 Mode 2 Carrier frequency 1.26e+009Hz 1.26e+009Hz Central incident angle 31.2851° 31.2911° Orbital azimuth 192.477621° 192.4766964° Distance pixel size 1.665514m 1.665514m Azimuth pixel size 1.281204m 3.046507m

[0103] Step 2: Radar Image CR Recognition

[0104] According to the on-site and previously known landslide sliding conditions, two sets of CR ( Figure 7 The 2023 / 06 / 27 image (mode 1) in the LT-1 satellite radar data was selected as the SAR main image, and the geographic coordinates recorded by the GNSS device were converted into the range Doppler coordinate framework of the 2023 / 06 / 27 image (SAR main image) according to formula (1). The row and column numbers of CR on the 2023 / 06 / 27 image were preliminarily identified, and finally confirmed by visual interpretation. Figure 3 The bright white square points in the figure are CR points.

[0105] Step 3: Multi-resolution InSAR deformation time series analysis of large deformation landslide

[0106] (1) Using the 2023 / 06 / 27 image parameter file, the satellite orbit azimuth covering the Xieliupao landslide area was calculated to be 192.4786°, and the incident angle was 31.2672°. According to formula (2), the GNSS observations at CR-1 were converted into a radar line-of-sight deformation time series, and the GNSS radar line-of-sight deformation rate was obtained to be approximately 1.1 cm / d. According to formulas (3) and (4), it was calculated that the monitoring of the middle and lower parts of the Xieliupao landslide required a 10-day LT-1 radar satellite revisit period. All radar images are registered to the range Doppler coordinates of the 2023 / 06 / 27 image (the registration accuracy is better than 0.1 pixel). Since the deformation rate of the rear edge and the middle of the landslide exceeds the order of magnitude of the CR-1 position, the time interval threshold (i.e., the time baseline threshold in step S106) is set to 8 days. Images with a time interval of no more than 8 days are selected from the registered SAR images as interferometric image pairs, and 33 interferometric image pairs are obtained. Then, the initial line-of-sight deformation rate (i.e., the initial line-of-sight deformation rate in the aforementioned step S109) is calculated according to formula (5).

[0107] According to the GNSS observation data at CR-1 and formula (3), it is calculated that the revisit period for measuring the middle and lower parts of the landslide is about 10 days, but the deformation of the middle and rear parts of the landslide is larger, so a faster revisit period is required. Combined with the 4-day or 8-day revisit period of the LT-1 satellite, and considering that more data can be involved in the calculation, the time interval is set to 8 days.

[0108] (2) Since June 22, 2023, the LT-1 satellite has switched from the on-orbit test mode (mode 2) to mode 1 to carry out routine photography of the Chinese land area. The actual revisit period of the same orbit may exceed 8 days. For example, the time interval between the two adjacent periods of 2023 / 07 / 09 and 2023 / 08 / 10 is 32 days, resulting in discontinuity in the unwrapped phase. Therefore, when analyzing the InSAR deformation time series, the initial line-of-sight deformation rate calculated according to formula (5) is multiplied by 32 days to replace the original differential interferogram of 2023 / 07 / 09_2023 / 08 / 10. Therefore, based on the interferometric image pair set with a time interval of no more than 8 days and the adjacent image pairs with a time interval of more than 8 days, the following is constructed: Figure 4 The 35 interferometric image pairs shown in the figure are used to form a continuous deformation time series. The differential interferometric atlas used for time series analysis includes two parts: one is the differential interferometric atlas with a time interval of no more than 8 days (33 in total); the other is that during the satellite shooting process, there will be a revisit cycle of more than 8 days. In order to make the deformation analysis time continuous, adjacent image pairs with a time interval of more than 8 days are also involved in the calculation (2 in total). The differential interferometric atlas of adjacent images is calculated by multiplying the initial deformation rate by the time interval of adjacent images.

[0109] Then, the coherence coefficient threshold is set to perform differential interferometry processing on 35 interferometric image pairs. The multi-view ratio in the range and azimuth directions is 2:4. The 2m resolution DEM produced by the airborne LiDAR is used to simulate the interferometric phase and compensate the terrain phase to obtain the differential interferogram. The Goldstein-Werner filtering method is used to filter the differential interferogram to obtain the filtered differential interferogram (e.g. Figures 5a to 5c As shown in Figure 2), the spatial phase unwrapping of the filtered differential interferogram is performed using the minimum cost flow (MCF) algorithm. The amplitude dispersion index and the coherence coefficient threshold are used to comprehensively select the coherent point targets with stable SAR image spectrum characteristics, and it is ensured that there are CR points in the coherent point targets. Then, the deformation rate and deformation time series of all coherent point targets are calculated using formula (6), and the following is obtained: Figure 7 and Figures 8a to 8c The InSAR deformation rate diagram and deformation time series diagram of the Xieliupao landslide are shown.

[0110] Step 4: Mutual calibration between CR-InSAR and GNSS data

[0111] A total of 24 data sets were collected from March 15 to August 18, 2023. Therefore, the deformation time series of the CR point has 24 time nodes. The observation data corresponding to the radar shooting date is found from the GNSS measurement value at CR-2 converted into the line-of-sight deformation observation sequence, and the mutual difference between CR-InSAR (i.e., the InSAR measurement value at the CR point) and the GNSS converted into the line-of-sight deformation observation value is obtained, and the root mean square error is calculated using formula (7). Fig. 9 As shown in the figure, the calculated root mean square error is 7 mm. The trends of the CR-InSAR and GNSS observation sequences are very consistent, achieving the expected monitoring effect.

[0112] Through the description of the above implementation modes, those skilled in the art can clearly understand that the present invention can be implemented by combining software with a hardware platform. Based on such an understanding, all or part of the contribution of the technical solution of the present invention to the background technology can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation modes of the present invention or some parts of the implementation modes.

[0113] Correspondingly, an embodiment of the present invention further provides a computer-readable storage medium on which computer-readable instructions or programs are stored. When the computer-readable instructions or programs are executed by a processor, the processor performs the following operations: the operations include the steps included in the method described in any of the above embodiments, which are not repeated here. Among them, the storage medium may include: for example, a CD, a hard disk, a floppy disk, a flash memory, a magnetic tape, etc. Part or all of the above embodiments of the present invention can be implemented by a computer device, which includes the above storage medium and a processor, and the processor executes the computer-readable instructions in the storage medium to perform the steps, operations, and processing recorded in any of the above embodiments.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention is described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention. Therefore, the protection scope of the present invention shall be subject to the claims.

Claims

1. A landslide deformation time series monitoring method using synthetic aperture radar interferometry (InSAR) that integrates corner reflector CR and global navigation satellite system GNSS, the method comprising: Acquiring satellite radar data covering the landslide area, and selecting a SAR main image from the satellite radar data; Obtain data sent from at least one set of CR and GNSS equipment deployed in the area with significant landslide deformation, wherein at least one set of CR and GNSS equipment is deployed at the same location of the landslide for accuracy verification; Converting the geographic coordinates recorded by the GNSS device into range Doppler coordinates of the SAR main image, and identifying the position of the CR on the SAR main image; Calculating satellite orbit information covering the landslide area according to the parameter file of the SAR main image, wherein the satellite orbit information includes a satellite orbit azimuth and a radar incident angle; Converting GNSS equipment observation values ​​into radar line-of-sight observation values ​​according to the satellite orbit information; Determine a radar sight line deformation time series based on the radar sight line observation value, calculate a theoretical radar satellite shooting time interval according to the radar sight line deformation time series, and set a time baseline threshold; Performing sub-pixel registration on all images in the satellite radar data to the range Doppler coordinate system of the main image to obtain a registered SAR image; Selecting images whose time intervals do not exceed the time baseline threshold from the registered SAR images as interference image pairs, and performing differential interference processing on the interference image pairs to obtain a differential interference atlas; Calculating the initial line-of-sight deformation rate according to the differential interference atlas; For adjacent images whose time intervals in the registered SAR images exceed the time baseline threshold, determining a replacement differential interferogram of the adjacent images according to the initial line-of-sight deformation rate and the time intervals of the adjacent images; The differential interference atlas set and the alternative differential interference atlas are used as standard data input for the interference point target analysis method IPTA, and coherent point targets including CR points are selected. The differential interference atlas set and the alternative differential interference atlas are subjected to deformation timing analysis to obtain the deformation rate and deformation time series of the coherent point targets.

2. The method according to claim 1, characterized in that The satellite radar data are different mode data covering the same landslide area, and the incident angles of the different mode data are basically consistent.

3. The method according to claim 1, characterized in that According to the distance equation R = |R s -R t |, and the Doppler equation Converting the geographic coordinates recorded by the GNSS device into range Doppler coordinates of the SAR main image; Where R is the distance from the sensor to the point target, R s is the position of the sensor, v s is the velocity vector of the sensor, R t is the position of the point target, v t is the velocity vector of the point target, λ is the radar wavelength, f d is the Doppler frequency.

4. The method according to claim 1, characterized in that The GNSS device observation value is converted into the radar line of sight observation value according to the following calculation formula: [Unsinα-Uecosα]sinθ+Uucosθ+δ los =g los Among them, Ue, Un, and Uu are the east-west, north-south, and vertical deformation components of the GNSS equipment, α is the satellite orbit azimuth, θ is the radar incident angle, and g is the los The radar line of sight observation value is obtained by converting the GNSS equipment observation value.

5. The method according to claim 4, characterized in that Calculating the theoretical radar satellite shooting time interval according to the radar line of sight deformation time series includes: The radar line of sight deformation rate is calculated according to the following formula: Among them, t0 is the starting observation time, t i is the observation cut-off time, Δv glos From the start observation time t0 to the end observation time t i The radar line of sight deformation rate between them; Δv glos Substitute Δv into the following calculation formula to calculate the theoretical radar satellite shooting time interval: Among them, Δv is the change value of the radar line of sight deformation rate in the same pixel, λ is the radar wavelength, and Δt is the theoretical radar satellite shooting time interval.

6. The method according to claim 1, characterized in that The initial line-of-sight deformation rate is calculated according to the following deformation rate calculation formula: Among them, Δt k is the time interval of the kth interference image pair in the differential interference atlas, φ k is the unwrapping phase of the kth group of interference image pairs in the differential interference atlas, N is the total value of the interference image pairs, and p is the initial line-of-sight deformation rate.

7. The method according to claim 6, characterized in that Determining the replacement differential interference pattern of the adjacent images according to the initial line-of-sight deformation rate and the time interval of the adjacent images comprises: The product of the initial sight line deformation rate and the time interval between adjacent images is the replacement sight line deformation amount of the adjacent images; A substitute differential interference pattern of the adjacent images is determined according to the substitute line-of-sight deformation amount.

8. The method according to claim 1, characterized in that The method further comprises: Using Goldstein-Werner filtering method to filter the differential interferogram set and the alternative differential interferogram; The spatial phase unwrapping of the filtered differential interferogram is performed using the minimum cost flow algorithm; The coherent point target with stable SAR image spectrum characteristics is comprehensively selected by using amplitude dispersion index and coherence coefficient threshold; The deformation rate and deformation time series of all coherent point targets are calculated according to the following formula: in, is the phase difference between the coherent point targets i and j in the kth interference image pair, θ is the incident angle, is the vertical baseline, t k is the time interval, is the relative elevation difference between neighboring targets i and j, is the relative deformation rate between neighboring targets i and j, is the other phase, λ is the radar wavelength, and R is the distance from the sensor to the point target; Aiming at the deformation rate and deformation time series of the coherent point target, the interference point target analysis method IPTA is used to perform deformation time series analysis.

9. The method according to claim 1, characterized in that The method further comprises: The root mean square error σ of the difference between the InSAR measurement value at the CR point and the GNSS-converted line-of-sight deformation observation value in the same time period is calculated according to the following formula: Among them, x i is the InSAR measurement value at point CR, y i is the deformation observation value converted from GNSS to line-of-sight direction at the CR point, and n is the number of statistical samples; The measurement accuracy of the InSAR results is judged by the root mean square error of the difference between the InSAR measurement value at the CR point and the deformation observation value converted into the line of sight by GNSS.

10. A computer-readable storage medium storing computer-readable instructions, characterized in that: The computer readable instructions are executed by a processor to implement the method according to any one of claims 1 to 9.