Three-dimensional deformation estimation method based on InSAR variance constraint elastic network regularization

Abstract

The invention discloses a three-dimensional deformation estimation method based on InSAR variance constraint elastic network regularization. The method comprises the steps that LOS direction deformation and azimuth direction deformation of each track are obtained; calculating the variance of all deformation pixels in the LOS direction observation area; calculating variances of all deformation pixels in the azimuth observation area, and establishing a three-dimensional deformation relation equation set; a first three-dimensional deformation calculation model is obtained; obtaining a three-dimensional direction pixel-level variance; obtaining a second three-dimensional deformation calculation model; calculating to obtain a three-dimensional direction pixel-level variance of the second three-dimensional deformation calculation model; establishing a third three-dimensional deformation calculation model; selecting a pixel-level regularization parameter of a third three-dimensional deformation calculation model in the Burst overlapping region; obtaining a regularization parameter resolving model of three-dimensional direction deformation; an improved third three-dimensional deformation calculation model is obtained; and extracting high-precision three-dimensional deformation according to the improved third three-dimensional deformation calculation model. And the purpose of extracting high-precision three-dimensional deformation is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of microwave remote sensing measurement, and in particular relates to a three-dimensional deformation estimation method based on InSAR variance constraint elastic net regularization. Background Art

[0002] Synthetic aperture radar interferometry (InSAR) technology has been an effective method for monitoring geological disasters such as landslides, earthquakes, volcanic eruptions, and mining subsidence in the past two decades. InSAR technology can provide the deformation in the line of sight (LOS) direction of the SAR satellite radar. Multiple aperture SAR interferometry (MAI) can provide the deformation in the azimuth direction. By combining the LOS direction and azimuth direction deformations of multiple orbits, the three-dimensional deformation of the disaster area can be solved. Since the monitoring accuracy of the azimuth direction deformation requires high Doppler bandwidth of the observation data and high coherence of the interferometric phase, in practical applications, only the SAR satellites in the TOPS working mode meet this requirement. Using the burst overlap interferometry (BOI) technology, the high-precision azimuth direction deformation of sparse pixel points within the burst overlap area can be obtained, but this technology cannot obtain the azimuth direction deformation of all pixel points within the observation area, resulting in problems such as missing three-dimensional deformation and low accuracy in the solution. In recent years, with the increase in the number of SAR satellites, it has become possible to obtain the LOS direction deformations of multiple orbits with different headings and incident angles (ascending orbit, descending orbit, left-looking, and right-looking data) in the same research area. However, the observation coefficient matrix constructed using the LOS direction deformations of multiple orbits has a certain degree of ill-conditioning problem, resulting in serious random errors and systematic errors in the solved three-dimensional deformation, especially in the vertical and north-south directions, both in time and space. Traditional methods cannot effectively suppress these two types of errors simultaneously, reducing the monitoring accuracy of the three-dimensional deformation and unable to meet the monitoring requirements of different disasters. Summary of the Invention

[0003] Aiming at the problems existing in the prior art, the present invention provides a three-dimensional deformation estimation method based on InSAR variance constraint elastic net regularization, which at least partially solves the problems of large random errors and systematic errors and low deformation monitoring accuracy existing in the prior art.

[0004] An embodiment of the present disclosure provides a three-dimensional deformation estimation method based on InSAR variance constraint elastic net regularization, including:

[0005] Obtaining the LOS direction deformation and azimuth direction deformation of each orbit based on multi-orbit SAR satellite data;

[0006] Calculate the LOS-directional residual of each orbit according to the LOS-directional deformation, and calculate the variance of all deformation pixels in the LOS-directional observation area based on the LOS-directional residual;

[0007] Calculate the variance of all deformation pixels in the azimuthal observation area according to the azimuthal interference phase coherence coefficient and the effective number of looks, where the azimuthal interference phase coherence coefficient and the effective number of looks are obtained based on multi-track SAR satellite data;

[0008] Establish the relationship equation set between the LOS-directional deformation and the three-dimensional deformation and the relationship equation set between the LOS-directional deformation, the azimuthal deformation and the three-dimensional deformation respectively;

[0009] According to the relationship equation set between the LOS-directional deformation and the three-dimensional deformation, use the weighted least squares method to obtain the first three-dimensional deformation solution model;

[0010] Input the variance of all deformation pixels in the LOS-directional observation area into the first error transfer model to calculate the three-dimensional direction pixel-level variance of the first three-dimensional deformation solution model;

[0011] According to the relationship equation set between the LOS-directional deformation, the azimuthal deformation and the three-dimensional deformation, use the weighted least squares method to obtain the second three-dimensional deformation solution model;

[0012] Input the variance of all deformation pixels in the LOS-directional observation area and the variance of all deformation pixels in the azimuthal observation area into the second error transfer model to calculate the three-dimensional direction pixel-level variance of the second three-dimensional deformation solution model;

[0013] Establish the third three-dimensional deformation solution model based on the elastic net regularization method;

[0014] Obtain the three-dimensional deformation in the Burst overlap area according to the second three-dimensional deformation solution model, divide the obtained three-dimensional deformation in the Burst overlap area into a test set and a training set, and select the pixel-level regularization parameter of the third three-dimensional deformation solution model in the Burst overlap area through cross-validation with the result obtained by the third three-dimensional deformation solution model;

[0015] Extract the three-dimensional direction pixel-level variance in the Burst overlap area based on the three-dimensional direction pixel-level variance, model the three-dimensional direction pixel-level variance in the Burst overlap area and the pixel-level regularization parameter in the Burst overlap area according to linear regression, and obtain the regularization parameter solution model of the three-dimensional direction deformation;

[0016] According to the regularization parameter solution model and the third three-dimensional deformation solution model, obtain the improved third three-dimensional deformation solution model;

[0017] Extract high-precision three-dimensional deformation according to the improved third three-dimensional deformation solution model.

[0018] The three-dimensional deformation estimation method based on InSAR variance constraint elastic network regularization provided by the present invention models the InSAR three-dimensional deformation variance and the elastic network regularization parameters through the high-precision three-dimensional deformation in the Burst overlapping area, obtains the regularization parameters of all pixel points in the observation area, and then uses the regularization parameters to solve the three-dimensional deformation, overcoming the random errors and systematic errors caused by matrix ill-conditioning and regularization parameter deviation, thereby improving the spatio-temporal three-dimensional deformation monitoring accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other objects, features, and advantages of the present disclosure will become more apparent by describing the exemplary embodiments of the present disclosure in more detail with reference to the accompanying drawings, wherein, in the exemplary embodiments of the present disclosure, the same reference numerals generally represent the same components.

[0020] Figure 1 It is a flowchart of the three-dimensional deformation estimation method based on InSAR variance constraint elastic network regularization provided by the embodiments of the present disclosure;

[0021] Figure 2 It is a schematic block diagram of an electronic device provided by the embodiments of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0023] It should be clear that the following illustrates the embodiments of the present disclosure through specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present disclosure without creative work belong to the scope of protection of the present disclosure.

[0024] Note that the following description relates to various aspects of embodiments within the scope of the appended claims. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is illustrative only. Based on this disclosure, those skilled in the art should 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 the aspects described herein can be used to implement a device and / or practice a method. Additionally, this device and / or method can be implemented using other structures and / or functionality in addition to one or more of the aspects described herein.

[0025] It should also be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present disclosure schematically. The diagrams only show the components related to the present disclosure and are not drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0026] In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the aspects can be practiced without these specific details.

[0027] Elastic net regularization combines L1 regularization and L2 regularization. By introducing L1 and L2 penalty terms, it overcomes the limitations of each of the two regularizations and can achieve the removal of random errors and systematic errors in InSAR deformation solution; the Burst overlap region belongs to the LOS direction or azimuth direction observation region.

[0028] For ease of understanding, as Figure 1 shown, this embodiment discloses a three-dimensional deformation estimation method based on InSAR variance-constrained elastic net regularization, including:

[0029] Step S101: Obtain the LOS-direction deformation and azimuth-direction deformation of each orbit based on multi-orbit SAR satellite data; specifically, obtain the LOS-direction deformation of each orbit from the multi-orbit SAR satellite data, specifically by processing the multi-orbit SAR satellite data using D-InSAR, MT-InSAR, or MAI technology to obtain the LOS-direction or azimuth-direction deformation of each orbit; then use the BOI technology for the multi-orbit SAR satellite data supporting the TOPS working mode to obtain the azimuth-direction deformation of each orbit.

[0030] Specifically, the BOI technology mainly performs three - differential interferences on the overlapping areas of Bursts (data blocks) in the primary and secondary SAR satellite data. The data of the primary and secondary overlapping areas both include the top - layer and bottom - layer overlapping areas. The first interference is performed on the top - layer overlapping areas of the primary and secondary data, the second interference is performed on the bottom - layer overlapping areas of the primary and secondary data, and the third interference is performed on the results of the first two interferences of the primary and secondary data. Then, phase - to - deformation conversion and geocoding processing are carried out to make the obtained azimuth - direction and LOS - direction deformation results in the same geographical coordinates. Among them, geocoding refers to converting the radar image coordinate system to the Universal Transverse Mercator (UTM) coordinate system.

[0031] Step S102: Calculate the LOS - direction residual of each orbit according to the LOS - direction deformation, and calculate the variance of all deformation pixels in the LOS - direction observation area based on the LOS - direction residual. Specifically: Calculate the LOS - direction residual of each orbit according to the obtained LOS - direction deformation, and calculate the variance of all deformation pixels in the LOS - direction observation area based on the LOS - direction residual of each orbit. The InSAR LOS - direction variance can be divided into two cases according to InSAR single - observation (i.e., DInSAR) and time - series observation (i.e., MT - InSAR). Specifically, when using DInSAR to obtain the interferogram of the observation area, the variance is calculated using the probability density function of the interference phase. The formula is:

[0032]

[0033] In the formula, μ 2 is the LOS - direction deformation pixel - level variance, λ is the wavelength, is the interference phase, is the expectation of the interference phase, is the probability density function of the interference phase, and its formula can be expressed as:

[0034]

[0035] Among them, γ is the coherence coefficient of the interference phase, Γ is the gamma function, L is the number of looks of the SAR image, and i is the size of the number of looks;

[0036] When using MT - InSAR to obtain the time - series interferogram of the observation area, the annual average deformation rate variance of each orbit can be obtained by processing the multi - source errors in the time - series interferogram through the MT - InSAR process flow to obtain the time - series residual of each orbit. The formula is:

[0037]

[0038] In the formula, μ 2 is the LOS - direction deformation pixel - level variance, σm is the MT-InSAR time series residual, and T m is the time interval between adjacent observations in the SAR satellite time series observation, is the average acquisition time interval, and M is the total number of SAR image observations in the same orbit.

[0039] Step S103: Calculate the variance of all deformation pixels in the azimuth observation area according to the azimuth interference phase coherence coefficient and the effective number of looks. The azimuth interference phase coherence coefficient and the effective number of looks are obtained based on multi-orbit SAR satellite data;

[0040] The azimuth interference phase coherence coefficient can be obtained by taking the average of the coherence coefficients of the forward and backward interferograms obtained by the MAI technique; the effective number of azimuth looks is the product of the azimuth looks, range looks, and normalized squint of the azimuth interferogram.

[0041] When using the MAI technique to obtain the azimuth interferogram of the observation area, the formula for calculating the variance of all deformation pixels in the azimuth observation area according to the azimuth interference phase coherence coefficient and the effective number of looks is expressed as:

[0042]

[0043] In the formula, ε 2 is the variance of the azimuth deformation pixels, M is the effective number of looks of the azimuth interference phase, and ω is the coherence coefficient of the azimuth interference phase.

[0044] Step S104: Establish a system of equations for the relationship between LOS deformation and three-dimensional deformation and a system of equations for the relationship between LOS deformation, azimuth deformation, and three-dimensional deformation. Specifically, according to the side-looking imaging geometry relationship of the SAR sensor, establish a system of equations for the relationship between LOS deformation and three-dimensional deformation, and the matrix expression is:

[0045]

[0046] The matrix expression of the system of equations for the relationship between LOS, azimuth deformation, and three-dimensional deformation is:

[0047]

[0048] In the formula, Dl i represents the LOS deformation of the i-th orbit, and Da i represents the azimuth deformation of the i-th orbit, where i = 1…n, and n is the number of orbits of different SAR satellites covering the same observation area.

[0049] θ i is the incident angle of the SAR radar electromagnetic wave, and α iis the clockwise angle between the due north direction and the flight direction of the SAR satellite. Du, De, and Dn are the three-dimensional deformations in the vertical, east-west, and north-south directions, respectively.

[0050] Step S105: According to the relationship equations of LOS direction deformation and three-dimensional deformation, use the weighted least squares method to obtain the first three-dimensional deformation solution model. Specifically, according to the relationship equations of LOS direction deformation and three-dimensional deformation, use the weighted least squares method to obtain the first three-dimensional deformation solution model as follows:

[0051]

[0052] In the formula, A1 is:

[0053] L1 is the deformation in the radar line-of-sight direction, which are respectively:

[0054] P1 is the weight matrix of multi-orbit LOS direction deformation,

[0055] D1 is the three-dimensional deformation obtained by solution,

[0056] Step S106: Input the variances of all deformation pixels in the LOS direction observation area into the first error propagation model to calculate the three-dimensional direction pixel-level variance. The first error propagation model can be expressed as:

[0057]

[0058] In the formula, V1 is the three-dimensional direction pixel-level variance, Y1 is the variance-covariance matrix of multi-orbit LOS direction, and Y1 is expressed as:

[0059]

[0060] In the formula, ρ i,j is the correlation between the variances of LOS direction deformations in different orbits, and μ i is the variance of the InSAR LOS direction of the i-th orbit.

[0061] Step S107: According to the relationship equations of LOS direction deformation, azimuth direction deformation and three-dimensional deformation, use the weighted least squares method to obtain the second three-dimensional deformation solution model;

[0062]

[0063] In the formula, A2 is:

[0064] P2 is the weight matrix of multi-orbit LOS direction and azimuth direction deformations.

[0065] D2 is the three-dimensional deformation, which are respectively:

[0066] Step S108: Input the variances of LOS in the observation area for all deformed pixels and the variances of azimuth in the observation area for all deformed pixels into the second error transfer model to calculate the three-dimensional directional pixel-level variance of the second three-dimensional deformation solution model. The second error transfer model is expressed as:

[0067]

[0068] In the formula, V2 is the three-dimensional directional pixel-level variance, and Y2 is the variance-covariance matrix of multi-orbit LOS and azimuth.

[0069] Y2 is expressed as:

[0070]

[0071] In the formula, κ i,j is the correlation between the azimuth deformation variances of different orbits, ξ i,j is the correlation between the LOS and azimuth deformation variances of different orbits, and ε i is the variance of the InSAR azimuth of the i-th orbit.

[0072] Step S109: Establish the third three-dimensional deformation solution model based on the elastic net regularization method:

[0073]

[0074] In the formula, D3 is the three-dimensional deformation, L3 ∈ {L1, L2}, A3 ∈ {A1, A2}, ||·||1 is the L1 regularization term, ||||2 is the L2 regularization term, υ is the regularization parameter, and η is the mixing parameter of the elastic net, which is used to balance the proportion of L1 and L2 regularization;

[0075] Take the derivative of the objective function and further organize to obtain the third three-dimensional deformation solution model as:

[0076]

[0077] In the formula, sgn(·) is the sign function, which is 1 when the variable in the function is greater than 0 and -1 when it is less than 0; max(·) is to obtain the maximum value of two variables; Y3 ∈ {Y1, Y2}; P3 ∈ {P1, P2}; H is the regularization matrix.

[0078] Step S110: Obtain the three-dimensional deformation of the Burst overlap area according to the second three-dimensional deformation solution model. Divide the obtained three-dimensional deformation of the Burst overlap area into a test set and a training set, and select the pixel-level regularization parameter of the third three-dimensional deformation solution model in the Burst overlap area through cross-validation with the result obtained by the third three-dimensional deformation solution model.

[0079] Step S111: Extract the three-dimensional directional pixel-level variance of the Burst overlapping region based on the three-dimensional directional pixel-level variance, and model the three-dimensional directional pixel-level variance of the Burst overlapping region and the pixel-level regularization parameter of the Burst overlapping region according to linear regression to obtain a regularization parameter solution model for three-dimensional directional deformation. Specifically, use linear regression to model the pixel-level variance of the Burst overlapping region and the pixel-level regularization parameter determined in step S110 to obtain a regularization parameter solution model for three-dimensional directional deformation. Specifically: Extract the three-dimensional directional pixel-level variance of the Burst overlapping region according to the three-dimensional directional pixel-level variance of all pixels in the observation region obtained in step S106 and / or step S108, and use the linear regression method to combine the three-dimensional directional pixel-level variance of the Burst overlapping region and the pixel-level regularization parameter of the Burst overlapping region obtained in step S110 to establish a regularization parameter solution model for all pixels in the observation region. The formula is:

[0080] υ = kο + b,

[0081] where k and b are parameters to be fitted by the linear regression method, and ο ∈ {μ 2 , ε 2}}.

[0082] Step S112: According to the regularization parameter solution model and the third three-dimensional deformation solution model, obtain the improved third three-dimensional deformation solution model as:

[0083]

[0084] Step S113: Extract high-precision three-dimensional deformation according to the improved third three-dimensional deformation solution model.

[0085] The electronic device disclosed in this embodiment includes a memory and a processor. The memory is used to store non-temporary computer-readable instructions. Specifically, the memory may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0086] The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In an embodiment of the present disclosure, the processor is used to run the computer-readable instructions stored in the memory, so that the electronic device executes all or part of the steps of the three-dimensional deformation estimation method based on InSAR variance constraint elastic network regularization in the foregoing embodiments of the present disclosure.

[0087] Those skilled in the art should understand that, in order to solve the technical problem of how to obtain a good user experience effect, this embodiment may also include well-known structures such as communication buses, interfaces, etc., and these well-known structures should also be included in the protection scope of the present disclosure.

[0088] As Figure 2 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. It shows a schematic structural diagram of an electronic device suitable for implementing the electronic device in the embodiments of the present disclosure. Figure 2 The shown electronic device is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present disclosure.

[0089] As Figure 2 As shown, the electronic device may include a processing device (such as a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) or the program loaded from the storage device into the random access memory (RAM). In the RAM, various programs and data required for the operation of the electronic device are also stored. The processing device, ROM, and RAM are connected to each other through a bus. The input / output (I / O) interface is also connected to the bus.

[0090] Generally, the following devices may be connected to the I / O interface: an input device including, for example, a sensor or a visual information acquisition device; an output device including, for example, a display screen; a storage device including, for example, a magnetic tape, a hard disk, etc.; and a communication device. The communication device may allow the electronic device to communicate wirelessly or wiredly with other devices (such as edge computing devices) to exchange data. Although Figure 2 the shown electronic device has various devices, it should be understood that it is not required to implement or have all the shown devices. Instead, more or fewer devices may be implemented or had.

[0091] In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present disclosure includes a computer program product that includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program code for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, all or part of the steps of the three-dimensional deformation estimation method based on InSAR variance constraint elastic network regularization according to the embodiments of the present disclosure are performed.

[0092] For a detailed description of this embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, and details will not be repeated here.

[0093] A computer-readable storage medium according to an embodiment of the present disclosure stores non-temporary computer-readable instructions. When the non-temporary computer-readable instructions are run by a processor, all or part of the steps of the multi-track InSAR high-precision three-dimensional deformation solution method according to the foregoing embodiments of the present disclosure are performed.

[0094] The above-mentioned computer-readable storage medium includes, but is not limited to: optical storage media (such as CD-ROMs and DVDs), magneto-optical storage media (such as MOs), magnetic storage media (such as magnetic tapes or external hard drives), media with built-in rewritable non-volatile memories (such as memory cards), and media with built-in ROMs (such as ROM cartridges).

[0095] For a detailed description of this embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, and details will not be repeated here.

[0096] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, benefits, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, benefits, effects, etc. are essential for each embodiment of the present disclosure. In addition, the above-mentioned specific details are only for illustrative and easy-to-understand purposes, rather than limitations, and the above details do not limit the present disclosure to necessarily adopt the above specific details for implementation.

[0097] In this disclosure, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. The block diagrams of devices, apparatuses, equipment, and systems involved in this disclosure are only illustrative examples and do not intend to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any way. Words such as "including", "comprising", "having", etc. are open-ended words, meaning "including but not limited to", and can be used interchangeably with each other. The words "or" and "and" used herein refer to the word "and / or", and can be used interchangeably with it, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to", and can be used interchangeably with it.

[0098] In addition, as used herein, "or" in the listing of items starting with "at least one" indicates a disjunctive listing, so that for example, the listing of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Further, the term "exemplary" does not mean that the examples described are preferred or better than other examples.

[0099] It should also be noted that in the systems and methods of this disclosure, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of this disclosure.

[0100] Various changes, substitutions, and alterations to the technologies described herein can be made without departing from the teachings defined by the appended claims. In addition, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and acts described above. Current or later-developed processes, machines, manufactures, compositions of events, means, methods, or acts that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or acts within their scope.

[0101] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

[0102] The foregoing description has been presented for purposes of illustration and description. In addition, this description is not intended to limit embodiments of the present disclosure to the form disclosed herein. Although several example aspects and embodiments have been discussed above, those of skill in the art will recognize some of their variations, modifications, alterations, additions, and subcombinations.

Claims

1. A three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization, characterized in that: include: Based on multi-orbit SAR satellite data, the LOS deformation and azimuth deformation of each orbit are obtained; The LOS residual of each track is calculated according to the LOS deformation, and the variance of all deformed pixels in the LOS observation area is calculated based on the LOS residual; Calculating the variance of all deformed pixels in the azimuth observation area according to the azimuth interferometric phase coherence coefficient and the effective multi-look number, wherein the azimuth interferometric phase coherence coefficient and the effective multi-look number are obtained based on multi-orbit SAR satellite data; Establish the relationship equations between LOS deformation and three-dimensional deformation and the relationship equations between LOS deformation, azimuth deformation and three-dimensional deformation respectively; According to the relationship equations between LOS deformation and 3D deformation, the first 3D deformation solution model is obtained by using weighted least square method. The variance of all deformed pixels in the observation area of ​​LOS is input into the first error transfer model to calculate the three-dimensional pixel-level variance; According to the relationship equations between LOS deformation, azimuth deformation and three-dimensional deformation, the second three-dimensional deformation solution model is obtained by using the weighted least square method. The variance of all deformed pixels in the LOS observation area and the variance of all deformed pixels in the azimuth observation area are input into the second error transfer model to calculate the three-dimensional direction pixel-level variance of the second three-dimensional deformation solution model; The third three-dimensional deformation solution model is established based on the elastic network regularization method; The three-dimensional deformation of the Burst overlapping area is obtained according to the second three-dimensional deformation solution model, and the obtained three-dimensional deformation of the Burst overlapping area is divided into a test set and a training set. The pixel-level regularization parameter of the third three-dimensional deformation solution model in the Burst overlapping area is selected by cross-validation with the results obtained by the third three-dimensional deformation solution model; Extracting the three-dimensional pixel-level variance of the Burst overlap region based on the three-dimensional pixel-level variance, modeling the three-dimensional pixel-level variance of the Burst overlap region and the pixel-level regularization parameter of the Burst overlap region according to linear regression, and obtaining a regularization parameter solution model for three-dimensional deformation; According to the regularization parameter solution model and the third three-dimensional deformation solution model, an improved third three-dimensional deformation solution model is obtained; High-precision 3D deformation is extracted based on the improved third 3D deformation solution model.

2. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 1, characterized in that: The calculation of the variance of all deformed pixels in the LOS observation area based on the LOS residual includes, when the D-InSAR technology is used to obtain the interference pattern of the observation area, the variance is calculated using the probability density function of the interference phase. The probability density function formula can be expressed as: In the formula, μ 2 is the pixel-level variance of the LOS deformation, λ is the wavelength, is the interference phase, is the expected interference phase, is the probability density function of the interference phase, and its formula can be expressed as: Where γ is the coherence coefficient of the LOS interference phase, Γ is the gamma function, is the interferometric phase after multiple views, L is the number of views of the SAR image, and i is the size of the number of views; When MT-InSAR is used to obtain the time-series interferogram of the observation area, the multi-source errors in the time-series interferogram are processed by MT-InSAR process processing to obtain the time-series residuals of each orbit and obtain the annual average deformation rate variance of each orbit, which is expressed as follows: In the formula, μ 2 is the pixel-level variance of the LOS deformation, σ m is the MT-InSAR time series residual, T m is the time interval between adjacent observations in SAR satellite time series observations, is the average acquisition time interval, M is the total number of SAR image observations in the same orbit; The variance of all deformed pixels in the LOS observation area includes the annual average deformation rate variance of each orbit.

3. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 2, characterized in that: The calculation of the variance of all deformed pixels in the azimuth observation area according to the azimuth interference phase coherence coefficient and the effective multi-view number includes: When the MAI technique is used to obtain the azimuth interference pattern of the observation area, the formula for calculating the variance of all deformed pixels in the azimuth observation area based on the azimuth interference phase coherence coefficient and the effective multi-view number is expressed as: In the formula, ε 2 is the variance of the azimuthally deformed pixels, M is the effective number of views of the azimuthally interferometric phase, and ω is the coherence coefficient of the azimuthally interferometric phase.

4. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 3 is characterized in that: The method of establishing a group of equations for the relationship between the LOS deformation and the three-dimensional deformation according to the geometric relationship of the SAR sensor side imaging includes establishing a group of equations for the relationship between the LOS deformation and the three-dimensional deformation according to the geometric relationship of the SAR sensor side imaging, and the matrix form of the equation group is: The relationship equations between the LOS deformation, the azimuth deformation and the three-dimensional deformation are established, and the matrix form of the equations is: Where Dl i represents the LOS deformation of orbit i, Da i represents the azimuthal deformation of the i-th orbit, i=1…nn is the number of orbits of different SAR satellites covering the same observation area; θ i is the incident angle of SAR radar electromagnetic wave, α i is the clockwise angle between the true north and the flight direction of the SAR satellite. Du, De, and Dn are the three-dimensional deformations in the vertical, east-west, and north-south directions, respectively.

5. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 4, characterized in that: According to the relationship equations between the LOS deformation and the three-dimensional deformation, the weighted least square method is used to obtain the first three-dimensional deformation solution model, which is expressed as follows: Where A1 is: L1 is the radar line of sight deformation, which are: P1 is the multi-orbit LOS deformation weight matrix, D1 is the three-dimensional deformation obtained by calculation, 6. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 5, characterized in that: The variance of all deformed pixels in the LOS observation area is input into the first error transfer model to calculate the three-dimensional pixel level variance. The first error transfer model is expressed as: Where V1 is the pixel-level variance in the three-dimensional direction, Y1 is the variance-covariance matrix of the multi-orbit LOS direction, Y2 is the variance-covariance matrix of the multi-orbit LOS direction and azimuth direction, and Y1 is expressed as: Where ρ i,j is the correlation between the LOS deformation variances of different orbits, μ i is the variance of the InSAR LOS direction of the ith track; The variance of all deformed pixels in the LOS observation area and the variance of all deformed pixels in the azimuth observation area are input into the second error transfer model to calculate the three-dimensional pixel-level variance of the second three-dimensional deformation solution model. The second error transfer model is expressed as: Where V2 is the pixel-level variance in the three-dimensional direction, Y2 is the variance-covariance matrix of the multi-orbit LOS direction and azimuth direction, Y2 is expressed as: Where κ i,j is the correlation between the deformation variances in different orbital azimuths, ξ i,j is the correlation between the deformation variances in the LOS and azimuth directions of different orbits, ε i is the variance of the InSAR azimuth of the ith track.

7. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 6, characterized in that: According to the relationship equations between the LOS deformation, the azimuth deformation and the three-dimensional deformation, the weighted least square method is used to obtain the second three-dimensional deformation solution model, which is expressed as follows: Where A2 is: P2 is the multi-orbit LOS and azimuth deformation weight matrix; D2 is the three-dimensional deformation, which are:

8. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 7, characterized in that: Based on the elastic network regularization method, the third three-dimensional deformation solution model is established. The formula of the third three-dimensional deformation solution model is expressed as follows: Where D3 is the three-dimensional deformation, L3∈{L1,L2}, A3∈{A1,A2}, ||·||1 is the L1 regularization term, ||||2 is the L2 regularization term, υ is the regularization parameter, and η is the mixing parameter of the elastic network, which is used to balance the ratio of L1 and L2 regularization; By taking the derivative of the objective function, the third three-dimensional deformation solution model is further sorted out as follows: Where sgn(·) is a sign function, when the variable in the function is greater than 0, it is 1, and when it is less than 0, it is -1; max(·) is to find the maximum value of two variables; Y3∈{Y1,Y2}; P3∈{P1,P2}; H is the regularization matrix.

9. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 8, characterized in that: The three-dimensional pixel-level variance of the Burst overlapping area and the pixel-level regularization parameter of the Burst overlapping area are modeled according to linear regression to obtain a regularization parameter solution model for three-dimensional directional deformation; comprising: extracting the three-dimensional pixel-level variance of the Burst overlapping area according to the three-dimensional pixel-level variance, and using a linear regression method to combine the three-dimensional pixel-level variance of the Burst overlapping area and the pixel-level regularization parameter of the Burst overlapping area to model to obtain a regularization parameter solution model for three-dimensional directional deformation, The regularization parameter solution model formula for three-dimensional directional deformation is: υ=kο+b, Where k and b are the parameters that need to be fitted by the linear regression method, ο∈{μ 2 , ε 2 }.

10. The three-dimensional deformation estimation method based on InSAR variance constrained elastic network regularization according to claim 9, characterized in that: According to the regularization parameter solution model and the third three-dimensional deformation solution model, the improved third three-dimensional deformation solution model is obtained as follows:

Images