Coastal tidal flat deformation monitoring method

Abstract

The invention discloses a coastal tidal flat deformation monitoring method, and aims to solve the problem that a PS-InSAR technology belongs to a high-coherence scattering point target, is suitable for an urban area or other areas with relatively stable interference conditions, and cannot perform coastal tidal flat ground surface deformation well. The method comprises the following steps of: registering the rest radar images with a main image, sampling to a pixel space of the main image, generating a differential interference image pair, calculating an amplitude deviation index, screening a PS point with a stable scattering characteristic as a candidate high-coherence ground reference point of an SBAS-InSAR technology, and increasing the signal-to-noise ratio of data through multi-view and filtering processing so as to obtain a high-coherence ground reference point of the SBAS-InSAR technology. The method comprises the following steps of: firstly, carrying out high-pass filtering and low-pass filtering to estimate a deformation rate and a residual phase, removing the atmospheric phase by utilizing high-pass filtering and low-pass filtering, and finally, connecting a plurality of short baselines through SVD (Singular Value Decomposition), solving a minimum norm and a least square solution of the short baselines, and obtaining a deformation rate and an accumulated deformation quantity of a research area so as to monitor the tidal flat ground surface deformation.

Description

Technical Field

[0001] The present invention relates to the field of deformation monitoring, and in particular to a method for monitoring coastal tidal flat deformation, and more specifically to a method for monitoring stability of coastal tidal flat deformation. Background Art

[0002] There are two main problems in coastal tidal flat deformation monitoring: (1) The periodic rise and fall of tides causes the tidal flat to alternately become submerged and exposed, which results in differences in the data obtained by the monitoring equipment at different times, affecting the continuity and accuracy of deformation monitoring. (2) Coastal tidal flats usually have complex topographic features, such as sandbars, shoals, and estuaries. These topographic features not only increase the difficulty of laying out the monitoring equipment, but may also lead to uneven data collection. For example, patent No. CN202411783402.3 determines N groups of state sequences based on the feature sequences, rainfall sequences, and tide height sequences of N data clusters, and inputs each group of state sequences into different trained prediction models to obtain N deformation sequences; based on the N deformation sequences, the future deformation of the shallow stratum at the target tidal flat monitoring point is determined. Patent No. CN202410798530.9 is based on time-series synthetic aperture radar (SAR) remote sensing images, making full use of time-series remote sensing observations to obtain tidal flat inundation information; and calculates the tidal flat remote sensing inundation frequency by weighted scaling based on the number of remote sensing observations, thereby improving the inversion accuracy of the tidal flat remote sensing inundation frequency; then, based on the mathematical definition of inundation frequency, a tidal flat topography inversion model based on the complementary cumulative distribution function of tide level is constructed to complete the tidal flat topography inversion of the target area.

[0003] To this end, the present invention statistically analyzes the amplitude and phase information of all images to detect a portion of ground features that maintain high coherence regardless of time and space as PS points. The interferometric phase of the PS points is unwrapped to obtain true phase change information. Utilizing SBASInSAR technology, a subset of SAR images with a smaller spatiotemporal baseline is selected for interferometric processing to generate several pairs of highly coherent interferograms. The phase information of the PS points is introduced as a constraint into the SBASInSAR matrix equation to improve the accuracy of deformation measurement. By solving the matrix equation, the deformation rate of each pixel is obtained, and combined with the high-precision deformation information of the PS points, the surface deformation information of the coastal tidal flat is obtained. Summary of the Invention

[0004] PS-InSAR technology is a highly coherent scattering point target and is suitable for urban areas or other areas with relatively stable interference conditions, but it cannot effectively monitor coastal tidal flat surface deformation. In this method, a suitable master image is selected from the multi-view data, taking into account factors such as temporal and spatial incoherence. The remaining radar images are registered with the master image and sampled into the pixel space of the master image to generate differential interferometric image pairs. The amplitude deviation index is then calculated and PS points with stable scattering characteristics are selected as candidate high-coherence ground reference points for the SBAS-InSAR technology. The signal-to-noise ratio of the data is increased through multi-look and filtering processing. Deformation rate and residual phase are estimated through operations such as re-flattening and atmospheric filtering. The atmospheric phase is removed using high-pass and low-pass filtering. Finally, multiple short baselines are connected through SVD to obtain their minimum norm and least squares solutions. The deformation rate and cumulative deformation of the study area are obtained to achieve monitoring of tidal flat surface deformation.

[0005] A method for monitoring coastal tidal flat deformation, characterized by comprising the following steps:

[0006] N+1 SAR images of different time periods were obtained in the study area, and k differential interferograms were obtained by using the set time and space thresholds. Differential interferometry phase It can be expressed as:

[0007] (1)

[0008] Where, represents the differential interferometry phase.

[0009] After removing the terrain residual phase, atmospheric delay phase, and various noise phases, the average rate v of ground subsidence can be expressed as:

[0010] (2)

[0011] By constructing a matrix equation and using the least squares method or singular value decomposition (SVD) method, the deformation rate can be calculated. The matrix equation is in the form of:

[0012] (3)

[0013] Where, represents the differential interferometric phase vector.

[0014] The present invention is beneficial in that existing mainstream coastal tidal flat deformation monitoring methods fail to consider the stability of monitored surface deformation, especially in complex environments. The more factors in the coastal tidal flat deformation image, the greater the resulting deformation monitoring error. To fully extract surface deformation in coastal tidal flat areas, the present invention combines PS-InSAR and SBAS-InSAR technologies. PS points with stable scattering characteristics generated by PS-InSAR technology are used as candidate high-coherence ground reference points for SBAS-InSAR technology. SBAS-InSAR technology is then used to process radar images to monitor tidal flat surface deformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is the experimental process

[0016] Figure 2 is the surface deformation rate

[0017] Figure 3 is the accumulated deformation of the surface

[0018] Figure 4 is the filtered interference pattern

[0019] Figure 5 is the coherence coefficient graph

[0020] Figure 6 It is a disentanglement diagram

[0021] Figure 7 The distribution map of GCP points selected after track refinement and re-leveling

[0022] Figure 8 is the result of the surface deformation rate

[0023] Figure 9 is the result of the accumulated deformation of the surface DETAILED DESCRIPTION

[0024] The experiments and steps of the present invention are further described in detail below with reference to the accompanying drawings.

[0025] Combine Figure 1 , describing the experimental process of this embodiment:

[0026] S1: This implementation uses Sentinel-1A SAR Single Look Complex (SLC) image data in IW mode and ascending orbit. The data spans 95 images from July 2015 to December 2023. Specific image acquisition date parameters are shown in Table 1.

[0027] Sentinel-1 data acquisition date (orbit raising, orbit: 157)

[0028]

[0029] S2: Sentinel-2 satellite data were obtained from the U.S. Geological Survey (https: / / earthexplorer.usgs.gov / ) and the European Space Agency (ESA) websites. Since 2015 imagery covering the study area was unavailable, and only two 2016 images meeting the requirements were available with a close time interval, this study selected eight L2A and L1C product data from 2016 to 2023, with data concentrated in the winter months. Precise orbit data were obtained from the ESA information website (https: / / scihub.copernicus.eu / gnss / # / home / ).

[0030] S3: The deformation rate and cumulative deformation obtained after the second PS inversion are the final results. However, since the radar data is in the slant range coordinate system, it is necessary to convert it to a common geographic coordinate system. Using the downloaded high-precision SRTM DEM data as the reference projection parameters, a PS point vector file is generated. This file includes information such as the average deformation rate, geographic coordinates, and deformation at each time for each PS point, which serves as the control point for SBAS-InSAR processing.

[0031] S4: Combine Figure 4-6 The study area was divided into several sub-areas to minimize the area of water in the image. After generating the interferometric image pair, differential interferometric processing such as differential interferometric filtering and phase unwrapping was performed to obtain the interferogram. The flatland and terrain phases were removed using the downloaded high-precision DEM data as a reference. To improve the signal-to-noise ratio of the image, the ratio of range and azimuth views was set to 4:1. Phase unwrapping was then performed based on the Delaunay minimum cost flow method. This method is not affected by low-coherence pixels and is suitable for low-coherence areas such as water bodies and densely vegetated areas.

[0032] S5: Combine Figure 7 In order to remove the residual constant phase and the phase ramp error that still exists after unwrapping through orbit refinement and re-flattening, a certain number of GCP points with high coherence need to be selected. Due to the high subjectivity of manual selection, the PS points generated by the PS-InSAR technology during inversion basically meet the requirements of SBAS-InSAR for GCP points. Therefore, the PS points generated by the PS-InSAR technology are used as GCP candidate points in the orbit refinement and re-flattening steps of the SBAS-InSAR technology. Points with a deformation rate of 0 and stable scattering characteristics are selected as GCP points. A large number of GCP points are selected in each sub-area.

[0033] S6: Due to the need to remove atmospheric delay phase errors, both SBAS-InSAR inversions use the SVD method to estimate deformation rates and residual elevations for selected GCPs with stable scattering properties. The first inversion uses the Delaunay minimum cost flow method to perform secondary unwrapping of all interferogram pairs to optimize the interferogram. Because water absorbs most energy, a large correlation coefficient threshold results in most pixel values being output as NaN, preventing the required information from being obtained. After multiple experiments, the correlation coefficient threshold was set to 0.1 to obtain deformation rates and residual elevations. The second inversion removes the atmospheric phase and residual phase from the deformation rates generated by the first inversion. High-pass and low-pass filtering are used to remove the atmospheric phase, resulting in the final deformation values in the time series.

[0034] Table 2 Deformation rate statistics

[0035]

[0036] In summary, the present invention can improve the monitoring accuracy of coastal tidal flat deformation to a certain extent, making the coastal tidal flat deformation closer to the true value, and is more practical and valuable.

[0037] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0038] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for monitoring coastal tidal flat deformation, characterized in that: The steps include: By statistically analyzing the amplitude and phase information of all images, a subset of features that maintain high coherence regardless of time and space are detected as PS points. The interferometric phase of the PS points is unwrapped to obtain true phase change information. Utilizing SBASInSAR technology, a subset of SAR images with a small temporal and spatial baseline is selected for interferometric processing to generate several pairs of highly coherent interferograms. The phase information of the PS points is introduced as a constraint into the SBASInSAR matrix equation to improve the accuracy of deformation measurements. By solving the matrix equation, the deformation rate of each pixel is obtained, and combined with the high-precision deformation information of the PS points, the surface deformation information of the coastal tidal flat is ultimately obtained. N+1 SAR images of different time periods were obtained in the study area. K differential interferograms can be obtained by setting time and space thresholds. The kth image is in the azimuth distance pixel coordinate system. Differential interferometry phase It can be expressed as: (1) Where, represents the differential interferometry phase; After removing the terrain residual phase, atmospheric delay phase and various noise phases, the average rate of ground subsidence v can be expressed as: (2) The deformation rate can be obtained by constructing a matrix equation and using the least squares method or singular value decomposition (SVD) method; the matrix equation is in the form of: (3) Where, represents the differential interferometric phase vector.

Images