InSAR frozen soil deformation monitoring method based on U inspection and multi-level networking strategy

Abstract

The invention relates to the technical field of frozen soil monitoring, in particular to an InSAR frozen soil deformation monitoring method based on U test and a multilevel network construction strategy, which comprises the following steps: for a target frozen soil region, selecting PS candidate points through an amplitude deviation index and acquiring interference phases of the PS candidate points; u inspection is adopted to identify a homogeneous pixel set, an EMI method is adopted to carry out time sequence phase optimization processing, goodness of fit is adopted to measure the quality of an optimal phase estimation result, and DS candidate points and interference phases thereof are obtained; a multi-level network construction strategy is adopted, high-quality PS candidate points and DS candidate points construct a first-level resolving network, and the remaining candidate points gradually construct a second-level resolving network for parameter resolving; and separating frozen soil deformation from the unwrapping phase of each measurement point obtained from the multi-level network construction strategy by adopting joint parameter modeling. According to the invention, the precision of the deformation parameter calculation result of the frozen soil region and the spatial density of the measurement points are obviously improved, so that the frozen soil deformation monitoring quality is obviously improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of frozen soil monitoring, and particularly relates to an InSAR frozen soil deformation monitoring method based on U test and multi-level network strategy. BACKGROUND

[0002] Under the background of the continuous intensification of global climate warming trend, permafrost is experiencing a significant degradation process. Permafrost degradation is usually accompanied by significant surface deformation, which seriously threatens the safety of engineering infrastructure in permafrost regions. Therefore, accurately obtaining the surface deformation in permafrost regions can indirectly reflect the change characteristics of the permafrost state. In recent years, synthetic aperture radar interferometry (InSAR) technology has been widely used in permafrost deformation monitoring and active layer thickness inversion research due to its high precision, high spatial resolution and dynamic deformation acquisition capability. Existing time series InSAR technology can be divided into two categories: permanent scatterer interferometry (PSI) and distributed scatterer interferometry (DSI). The former is aimed at phase-stable permanent scatterer targets (PS), which is very limited in natural scenes such as permafrost regions. In contrast, DSI is aimed at distributed scatterer targets (DS) widely distributed in natural scenes such as farmland, cultivated land, and bare land. Although the signal-to-noise ratio of DS targets is low, the surrounding pixels usually exhibit similar scattering characteristics, so that their phase quality can be enhanced through phase optimization processing. Therefore, fully exploiting the deformation information of DS targets is of great significance to improve the quality and spatial density of surface deformation monitoring results in permafrost regions.

[0003] Accurate selection of homogeneous points is the basis and key to ensuring the quality of phase optimization results in DSI technology. Existing homogeneous point selection methods can be divided into two categories: 1) non-parametric hypothesis testing method; 2) parametric hypothesis testing method. Among them, the parametric hypothesis testing method assumes that the sample probability distribution is known, and tests whether the distribution parameters of two samples are the same by constructing appropriate statistical quantities. Compared with the non-parametric hypothesis testing method, it greatly improves the efficiency and efficacy of point selection, and is widely used. However, this method usually assumes that the backscattering characteristics of ground objects do not change over time. However, in permafrost regions, the soil moisture of the active layer will change periodically with temperature under the influence of the freezing and thawing process, which will cause the backscattering properties of the ground surface to also change significantly periodically. Therefore, the existing homogeneous point selection method based on parametric hypothesis testing will no longer be applicable in permafrost regions, and the time-varying characteristics of the ground surface backscattering need to be coupled in the hypothesis testing model.

[0004] In addition, in order to balance the calculation accuracy and efficiency in large-scale InSAR deformation monitoring processing, the existing research usually adopts the strategy of block and hierarchical processing. The strategy divides the research area into multiple spatial grids that overlap with each other, constructs a primary network with high-quality permanent scatterer targets, and establishes a secondary network by connecting the surrounding distributed scatterer targets. However, in permafrost regions, the number of permanent scatterer targets is small and their spatial distribution is extremely uneven. The traditional block and hierarchical processing strategy can easily make the solving network unstable or even appear as an island. At the same time, this can further cause a large number of distributed scatterer targets to be incorrectly excluded due to the lack of reliable reference points around them, thereby seriously affecting the number of final measurement points and the quality of the deformation results. Therefore, it is urgent to develop a network solving strategy applicable to severe incoherence environments to ensure the integrity and reliability of the deformation field solving in permafrost regions.

[0005] Based on the above analysis of the existing InSAR permafrost deformation monitoring methods, it can be seen that although InSAR technology has been widely used in permafrost deformation monitoring, it does not consider the time-varying characteristics of the backscattering of the ground surface in permafrost regions, and the deformation results are severely affected by incoherence, which can affect the accuracy and reliability of the InSAR deformation monitoring results in permafrost regions. SUMMARY

[0006] The purpose of the present application is to solve the problems of the existing homogeneous point selection method based on parameter hypothesis testing being unsuitable for permafrost regions and the existing block and hierarchical processing strategy having poor application effect in permafrost regions, in order to further improve the quality of permafrost deformation monitoring.

[0007] In order to achieve the above purpose, the present application provides an InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy, characterized in that it comprises the following steps:

[0008] S1, calculating the amplitude dispersion index of the SAR amplitude data set of the target permafrost region, setting the amplitude dispersion index threshold, selecting the PS candidate points and obtaining their interference phases;

[0009] S2, identifying the homogeneous pixel set using U-test on the SAR intensity data set of the target permafrost region, performing time series phase optimization processing using the EMI method, measuring the quality of the optimal phase estimation result using the goodness of fit, obtaining the DS candidate points and their interference phases;

[0010] S3, using a multi-level network construction strategy to construct a primary solving network corresponding to the PS candidate points and the DS candidate points based on the pre-set goodness of fit, and uniformly dividing multiple groups according to the goodness of fit from high to low, and gradually constructing secondary solving networks for parameter solving;

[0011] S4 uses joint parameter modeling to separate permafrost deformation from the unwrapped phase of each measurement point obtained by a multi-level network strategy.

[0012] Furthermore, S1 includes the following sub-steps:

[0013] The SAR image dataset of the target permafrost region is registered; appropriate temporal and spatial baseline thresholds are selected to generate small baseline interferograms; differential interferometry is performed on the small baseline interferograms using external digital elevation model data to obtain small baseline differential interferograms; based on the row and column information of PS candidate points that meet the amplitude deviation index threshold condition, the small baseline interferometric phase of the PS candidate points is extracted from the small baseline differential interferograms.

[0014] Furthermore, the amplitude deviation index in S1 The calculation formula is:

[0015]

[0016] in, and Represent the standard deviation and mean of the pixel amplitude time series, respectively; set Threshold, when the pixel Points smaller than the threshold are selected as PS candidate points.

[0017] Furthermore, S2 includes the following sub-steps:

[0018] S21, Register the SAR image dataset of the target permafrost region;

[0019] S22, Calculate SAR intensity data based on SAR image dataset, and obtain homogeneous pixel set for each point based on U test;

[0020] S23, Select the common master image and generate a single master image differential interferogram;

[0021] S24. Based on the homogeneous pixel set of each point, construct the complex coherence matrix, reconstruct the optimal phase that satisfies phase consistency using the EMI method, obtain the optimized single master image differential interferogram, and use the goodness of fit to measure the quality of the optimal phase estimation result.

[0022] S25. Based on the optimized single master image differential interferogram, obtain the small baseline differential interferogram;

[0023] S26. Based on the small baseline differential interferogram, extract the small baseline interferometric phase from the small baseline differential interferogram according to the row and column numbers of the DS candidate points.

[0024] Furthermore, the EMI method in S24 estimates a set of optimal single-master image interferometric phases that satisfy phase consistency from the complex coherence matrix through eigenvalue decomposition, thereby optimizing the phase value at each point.

[0025] Furthermore, in S24, a goodness-of-fit threshold is set, and pixels with a goodness-of-fit greater than the threshold are selected as DS candidate points and their interference phase is extracted.

[0026] Furthermore, S3 includes the following sub-steps:

[0027] S31, Construct a first-level solution network, and construct observation equations for the differential phase observations of each arc segment in the first-level solution network;

[0028] S32, the M-estimation method is used to solve for the arc segment parameters;

[0029] S33, calculate the equivalent time-domain coherence coefficient based on the arc segment residual, select and retain arc segments with an equivalent time-domain coherence coefficient greater than a preset threshold;

[0030] S34, perform closed loop detection on the spatial domain triangular network composed of the retained arc segments, and remove arc segments with phase ambiguity;

[0031] S35, set deformation reference points, use weighted least squares method to unwrap phase in the spatial domain, obtain the unwrapped phase of each measurement point, and complete the solution of the first-level network;

[0032] S36, an adaptive method is used to expand the remaining measurement points through a multi-level network. Based on the goodness of fit, the remaining measurement points are sorted from high to low and evenly divided into... Groups are processed sequentially, and the measurement points within each group are compared with their surroundings. A local star network is constructed using the nearest neighbor reference points, and the parameters are calculated. The retained points are added to the first-level network and used as references for subsequent network calculations.

[0033] Furthermore, in S4, when modeling the joint parameters, the components in the unwrapped phase of each measurement point obtained in S3 are modeled.

[0034] Furthermore, the unwrapping phase includes the permafrost deformation phase, the atmospheric delay phase, and the DEM error phase.

[0035] Furthermore, based on existing frozen soil deformation models, the frozen soil deformation phase is described using a linear + periodic deformation model, with the following formula:

[0036]

[0037] in, Indicates period, Indicates linear deformation rate. and Indicates seasonal deformation parameters. This represents a constant term.

[0038] The above-described solution of the present invention has the following beneficial effects:

[0039] The InSAR permafrost deformation monitoring method provided by this invention, based on the U-test and a multi-level meshing strategy, identifies homogeneous point sets for each pixel in the permafrost region through the U-test, taking into account the temporal variation characteristics of surface backscattering caused by the freeze-thaw process. This solves the problem that the traditional homogeneous point selection method based on parameter hypothesis testing is not applicable to permafrost regions. At the same time, when calculating deformation parameters by combining PS and DS candidate points, the method considers the problem of sparse and uneven distribution of PS points in the permafrost region and adopts a multi-level meshing strategy for deformation parameter calculation, which greatly improves the accuracy of deformation parameter calculation results and the spatial density of measurement points in the permafrost region, thereby significantly improving the quality of permafrost deformation monitoring.

[0040] Other beneficial effects of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0041] Figure 1 This is a flowchart of the steps of the present invention;

[0042] Figure 2 This is a graph showing the number of homogeneous points obtained by the U-test method in this embodiment of the invention;

[0043] Figure 3 This is an interferogram after phase optimization based on the EMI method in an embodiment of the present invention;

[0044] Figure 4 The image shows the InSAR deformation monitoring results obtained based on a multi-level network strategy in this embodiment of the invention, where (a) is the estimated linear deformation rate and (b) is the estimated seasonal deformation. Detailed Implementation

[0045] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0046] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0047] It should also be noted that the illustrations provided in the following embodiments are merely schematic representations of the basic concept of this disclosure. The illustrations only show components relevant to this disclosure and are not drawn according to the actual number, shape, and size of components in implementation. In actual implementation, the type, quantity, and proportion of each component can be arbitrarily changed, and the component layout may be more complex. Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0048] like Figure 1 As shown, embodiments of the present invention provide an InSAR permafrost deformation monitoring method based on the U-test and a multi-level meshing strategy. Addressing the current problem of low quality in InSAR permafrost deformation monitoring, this method primarily uses the U-test to identify homogeneous point sets for each pixel in the permafrost region, taking into account the temporal variation characteristics of surface backscattering caused by freeze-thaw processes. Furthermore, when calculating deformation parameters by combining PS and DS candidate points, the method considers the sparse and uneven distribution of PS points in the permafrost region and employs a multi-level meshing strategy for deformation calculation. This significantly improves the accuracy of the permafrost deformation calculation results and the spatial density of measurement points, thereby enhancing the quality of permafrost deformation monitoring. The method specifically includes the following steps:

[0049] S1: Calculate the amplitude deviation index for the SAR amplitude dataset of the target permafrost region, set the amplitude deviation index threshold, select PS candidate points and obtain their interference phase.

[0050] In this embodiment, the sub-steps for obtaining the interferometric phase of PS candidate points include: registering the SAR image dataset of the target permafrost region; selecting appropriate temporal and spatial baseline thresholds to generate a small baseline interferogram; performing differential interferometry on the small baseline interferogram using external digital elevation model (DEM) data to obtain a small baseline differential interferogram; and extracting the small baseline interferometric phase of the PS candidate points from the small baseline differential interferogram based on the row and column number information of the PS candidate points that meet the amplitude deviation index threshold condition.

[0051] Among them, the amplitude deviation index The calculation formula is:

[0052]

[0053] in, and These represent the standard deviation and mean of the pixel amplitude time series, respectively. By setting... The threshold is only applied when the pixel's Points below the threshold are selected as PS candidate points.

[0054] S2, the U test is used to identify homogeneous pixel sets in the SAR intensity dataset of the target permafrost area, and the EMI method is used for time-series phase optimization to obtain DS candidate points and their interference phases.

[0055] In this embodiment, this step specifically includes the following sub-steps:

[0056] S21, register the SAR image dataset of the target permafrost region.

[0057] S22, calculate SAR intensity data based on SAR image dataset, and obtain the set of homogeneous pixels for each point based on U test.

[0058] Specifically, when calculating SAR intensity, under the assumption of a complex circular Gaussian distribution, the single-look intensity... It follows an exponential distribution, and its natural logarithm is... Follow the mean The variance is The Fischer-Tippett distribution, where Let be the Euler constant. According to the product model, SAR image intensity is represented as a scene signal. With noise The product of, i.e. For two spatially adjacent pixels , Construct random variables And for The mean of multi-view SAR images The expected value is 0, and the variance is . .

[0059] Therefore, based on the calculated SAR intensity time series data, statistics are constructed and calculated. According to the central limit theorem, its time series mean It approximately follows a mean of 0 and a variance of . It follows a normal distribution. Given a significance level... The confidence interval estimate can be defined as follows:

[0060]

[0061] in, For standard normal distribution Quantiles. The statistical hypothesis testing window size is set, and the above statistic is constructed pixel-by-pixel with the pixel to be tested. If a pixel falls within the confidence interval, it will be identified as a homogeneous pixel of the pixel to be tested, thus obtaining a set of homogeneous pixels.

[0062] S23, Select the common master image and generate a single master image differential interferogram.

[0063] S24. Based on the homogeneous pixel set at each point, construct the complex coherence matrix and reconstruct the optimal phase that satisfies phase consistency using the EMI method. for:

[0064]

[0065] in, For coherence matrix, It represents the Hadamah accumulation. This indicates the conjugate transpose. The optimal phase to be estimated is denoted as .

[0066] It should be noted that the EMI method uses mathematical eigenvalue decomposition to estimate a set of optimal single-master image interferometric phases that satisfy phase consistency from the complex coherence matrix, thereby optimizing the phase value of each point and obtaining the optimized single-master image differential interferogram.

[0067] In this embodiment, a goodness-of-fit method is also employed. To measure the quality of the optimal phase estimation result, a goodness-of-fit threshold is set. Pixel-by-pixel inspection selects pixels larger than the threshold as candidate points for DS and extracts their interference phase. The goodness of fit is then considered. The calculation formula is:

[0068]

[0069] in, To take the real part of a complex number, The original interference phase, This is the optimized interference phase.

[0070] S25, based on the optimized single master image differential interferogram, obtain the small baseline differential interferogram similar to S1.

[0071] S26. Based on the small baseline differential interferogram, extract the small baseline interferometric phase from the small baseline differential interferogram according to the row and column numbers of the DS candidate points.

[0072] S3 employs a multi-level network construction strategy, building a first-level solution network from high-quality PS and DS candidate points, and then uniformly dividing the remaining candidate points according to their goodness of fit from high to low. The system is divided into groups, and secondary solution networks are gradually constructed to solve the parameters.

[0073] It should be noted that high-quality DS candidate points can be selected by setting a higher goodness-of-fit threshold in S2. There are various ways to construct the first-level solution network, such as using a locally free-connected network or a Delaunay triangular network.

[0074] In this embodiment, this step may specifically include the following sub-steps:

[0075] S31, Construct a first-level solution network, and build observation equations for the differential phase observations of each arc segment in the first-level solution network:

[0076]

[0077] in, Represents the observation vector. Represents the coefficient matrix. This represents the parameter to be determined.

[0078] S32, the M-estimation method is used to solve for the arc segment parameters. The method used in the [missing information] section... The parameter solution result for the next iteration is:

[0079]

[0080] in, The weight matrix is ​​initially set to the coherence of each interference phase, and then updated using the IGG weight function based on the residual magnitude.

[0081]

[0082] in, and These are constants, taking values ​​of 1.5 and 2.5 respectively. , For the observation residuals, .

[0083] S33, Calculate the equivalent time-domain coherence coefficient based on the arc segment residual. ,choose Arc segments larger than the specified threshold are retained.

[0084] S34, perform closed loop detection on the spatial domain triangular network composed of the retained arc segments, and remove arc segments that may have phase ambiguity.

[0085] S35. Set deformation reference points, use weighted least squares method to unwrap the spatial domain phase, obtain the unwrapped phase of each measurement point, and complete the solution of the first-level network.

[0086] S36, an adaptive method is used to expand the network at multiple levels for the remaining measurement points, based on the goodness of fit. The remaining measurement points are sorted from highest to lowest and divided evenly. Groups are processed sequentially, and the measurement points within each group are compared with their surroundings. A local star network is constructed using the nearest neighbor reference points, and the parameters are solved using the same method as the first-level network. The retained points are added to the first-level network and used as references for subsequent network solutions.

[0087] It should be noted that this process is iterative; points not retained at the end of the iteration will be added to the next group for reprocessing. Therefore, the multi-level network construction strategy adopted in this embodiment is suitable for network calculation in environments with severe decoherence, and can effectively ensure the integrity and reliability of the deformation field calculation in permafrost regions.

[0088] S4 uses joint parameter modeling to separate permafrost deformation from the unwrapped phase of each measurement point obtained by a multi-level network strategy.

[0089] Joint parameter modeling involves modeling the components of the unwrapped phase at each measurement point acquired by S3. Typically, the unwrapped phase mainly includes the permafrost deformation phase, atmospheric delay phase, and DEM error phase. The atmospheric delay phase primarily considers the residual tropospheric delay phase related to topography and can be modeled as a linear function with topography as the independent variable. The permafrost deformation phase, based on existing permafrost deformation models, adopts a linear + periodic deformation model, with the following formula:

[0090]

[0091] in, Indicates period, Indicates linear deformation rate. and Indicates seasonal deformation parameters. This represents a constant term.

[0092] Therefore, in this embodiment, the time-varying characteristics of surface backscattering in the permafrost region are considered when separating permafrost deformation, which further improves the accuracy and reliability of InSAR deformation monitoring results in the permafrost region.

[0093] The following case study further illustrates the effectiveness of this method. The estimated area is located in the southwest of a province, with an altitude ranging from 4330 to 5370 meters. The average annual temperature is approximately -3.8℃, and the average annual precipitation is approximately 300 mm. The surface cover is mainly alpine meadow, alpine grassland, and alpine desert, with permafrost and seasonal permafrost as the dominant permafrost types. The InSAR data used are Sentinel-1 rising orbit imagery, covering the period from October 2017 to May 2022.

[0094] In S1, Sentinel-1 images are registered to ensure a registration accuracy better than 1 / 1000 pixel; amplitude is calculated based on the registered SAR images to obtain SAR amplitude time series images, according to the formula... The amplitude deviation index was calculated, and a threshold of 0.25 was set. Pixels with an ADI less than this threshold were selected as PS candidate points. The time baseline and spatial baseline thresholds were set to 200m and 48 days, respectively. Interferometric pairs that met the threshold conditions were obtained, and differential interferometry was performed on the Space Shuttle Radar Topography Mission (SRTM) DEM data to obtain a series of small baseline differential interferograms. Based on the coordinate information of the selected PS candidate points, the small baseline differential interferometric phase of each PS point was extracted point by point.

[0095] In S2, the intensity of the registered SAR image is calculated to obtain a SAR intensity time series image. The window size is set to 43×11 (range × azimuth). The final result of the number of homogeneous points for each pixel is as follows. Figure 2 As shown.

[0096] Using the SAR image acquired on March 24, 2020, as the common master image, all other images were interferometrically processed with the master image to obtain a series of single master image interferograms. Then, differential interferometric processing was performed using SRTMDEM data to obtain a series of single master image differential interferograms. The optimized single master image differential interferograms obtained according to the sub-step corresponding to S2 are shown below. Figure 3 As shown. Based on the sub-step corresponding to S2, the corresponding small baseline differential interferogram is generated, and the small baseline interferometric phase of the DS candidate points is extracted from it.

[0097] In step S3, candidate DS points with a goodness-of-fit greater than 0.9 are selected as high-quality points and, together with candidate PS points, a first-level solution network is constructed and parameters are calculated. Simultaneously, the equivalent time-domain coherence coefficient is calculated based on the arc segment residuals, and arc segments with an equivalent time-domain coherence coefficient greater than 0.9 are retained. The multi-level network expansion and parameter calculation for the remaining measurement points are completed through the sub-steps corresponding to S3.

[0098] In S4, a linear + periodic deformation model is used to describe the deformation characteristics of permafrost. The final InSAR deformation results are as follows: Figure 4As shown. Therefore, the InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy provided in this embodiment identifies the homogeneous point set of each pixel in the permafrost area through U-test, taking into account the temporal variation characteristics of surface backscattering in the permafrost area caused by freeze-thaw process, and solves the problem that the traditional homogeneous point selection method based on parameter hypothesis testing is not applicable in the permafrost area; at the same time, when combining PS and DS candidate points to calculate deformation parameters, the problem of sparse and uneven distribution of PS points in the permafrost area is taken into account, and a multi-level network construction strategy is adopted for deformation parameter calculation, which greatly improves the accuracy of deformation parameter calculation results and the spatial density of measurement points in the permafrost area, thereby significantly improving the quality of permafrost deformation monitoring.

[0099] Based on the same inventive concept, this embodiment also provides an apparatus, including: a memory for storing a computer program; and a processor for executing the computer program to implement the relevant steps of the InSAR permafrost deformation monitoring method based on U-test and multi-level network strategy as described above.

[0100] The processor may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor can be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor may also include a main processor and coprocessors. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessors are low-power processors used to process data in the standby state. In some embodiments, the processor may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor may also include an Artificial Intelligence (AI) processor, which handles computational operations related to machine learning.

[0101] The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory is used to store at least the following computer program, which, after being loaded and executed by the processor, is capable of implementing the aforementioned software method steps. In addition, the resources stored in the memory may also include operating systems and data, and the storage method may be temporary or permanent storage. The operating system may include Windows, Unix, Linux, etc.

[0102] This embodiment also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the relevant steps of the InSAR permafrost deformation monitoring method based on U-test and multi-level network strategy as described above.

[0103] The apparatus and computer-readable storage medium provided in this embodiment have the same inventive concept and beneficial effects as the aforementioned methods, and will not be described again here.

[0104] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0105] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An InSAR method for monitoring permafrost deformation based on U-test and multi-level network construction strategy, characterized in that, Includes the following steps: S1, calculate the amplitude deviation index for the SAR amplitude dataset of the target permafrost region, set the amplitude deviation index threshold, select PS candidate points and obtain their interferometric phase; S2, the U test is used to identify homogeneous pixel sets in the SAR intensity dataset of the target permafrost area, the EMI method is used for time-series phase optimization, the goodness of fit is used to measure the quality of the optimal phase estimation result, and DS candidate points and their interference phases are obtained. S3 employs a multi-level network construction strategy, constructing a first-level solution network by combining the corresponding PS candidate points and the DS candidate points that meet the preset goodness-of-fit criteria based on S2. The remaining candidate points are evenly divided into multiple groups according to the goodness-of-fit from high to low, and secondary solution networks are gradually constructed to solve the parameters. S4 uses joint parameter modeling to separate permafrost deformation from the unwrapped phase of each measurement point obtained by a multi-level network strategy.

2. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 1, characterized in that, S1 includes the following sub-steps: The SAR image dataset of the target permafrost region is registered; appropriate temporal and spatial baseline thresholds are selected to generate small baseline interferograms; differential interferometry is performed on the small baseline interferograms using external digital elevation model data to obtain small baseline differential interferograms; based on the row and column information of PS candidate points that meet the amplitude deviation index threshold condition, the small baseline interferometric phase of the PS candidate points is extracted from the small baseline differential interferograms.

3. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 2, characterized in that, S1 Amplitude Deviation Index The calculation formula is: ； in, and Represent the standard deviation and mean of the pixel amplitude time series, respectively; set Threshold, when the pixel Points smaller than the threshold are selected as PS candidate points.

4. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 1, characterized in that, S2 includes the following sub-steps: S21, Register the SAR image dataset of the target permafrost region; S22, Calculate SAR intensity data based on SAR image dataset, and obtain homogeneous pixel set for each point based on U test; S23, Select the common master image and generate a single master image differential interferogram; S24. Based on the homogeneous pixel set of each point, construct the complex coherence matrix, reconstruct the optimal phase that satisfies phase consistency using the EMI method, obtain the optimized single master image differential interferogram, and use the goodness of fit to measure the quality of the optimal phase estimation result. S25. Based on the optimized single master image differential interferogram, obtain the small baseline differential interferogram; S26. Based on the small baseline differential interferogram, extract the small baseline interferometric phase from the small baseline differential interferogram according to the row and column numbers of the DS candidate points.

5. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 4, characterized in that, The EMI method in S24 estimates a set of optimal single-master image interferometric phases that satisfy phase consistency from the complex coherence matrix through eigenvalue decomposition, thus optimizing the phase value at each point.

6. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 4, characterized in that, In S24, a goodness-of-fit threshold is set, and pixels larger than the threshold are selected as DS candidate points and their interference phase is extracted.

7. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 1, characterized in that, S3 includes the following sub-steps: S31, Construct a first-level solution network, and construct observation equations for the differential phase observations of each arc segment in the first-level solution network; S32, the M-estimation method is used to solve for the arc segment parameters; S33, calculate the equivalent time-domain coherence coefficient based on the arc segment residual, select and retain arc segments with an equivalent time-domain coherence coefficient greater than a preset threshold; S34, perform closed loop detection on the spatial domain triangular network composed of the retained arc segments, and remove arc segments with phase ambiguity; S35, set deformation reference points, use weighted least squares method to unwrap phase in the spatial domain, obtain the unwrapped phase of each measurement point, and complete the solution of the first-level network; S36, an adaptive method is used to expand the remaining measurement points through a multi-level network. Based on the goodness of fit, the remaining measurement points are sorted from high to low and evenly divided into... Groups are processed sequentially, and the measurement points within each group are compared with their surroundings. A local star network is constructed using the nearest neighbor reference points, and the parameters are calculated. The retained points are added to the first-level network and used as references for subsequent network calculations.

8. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 1, characterized in that, When modeling the joint parameters in S4, the components of the unwrapped phase of each measurement point obtained in S3 are modeled.

9. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 8, characterized in that, The unwrapping phase includes the permafrost deformation phase, the atmospheric delay phase, and the DEM error phase.

10. The InSAR permafrost deformation monitoring method based on U-test and multi-level network construction strategy according to claim 9, characterized in that, The frozen soil deformation phase is described by existing frozen soil deformation models, using a linear + periodic deformation model, with the following formula: ； in, Indicates period, Indicates linear deformation rate. and Indicates seasonal deformation parameters. This represents a constant term.

Images