A method for atmospheric phase correction and deformation extraction of a distributed InSAR system

Abstract

The application discloses a distributed InSAR system atmospheric phase correction and deformation extraction method, belongs to the field of satellite image processing, and is characterized in that the method comprises the following steps: acquiring multi-satellite near-synchronous observation data of a distributed InSAR system to perform image registration; performing conjugate multiplication on the registered images to obtain interference phases; performing difference on the two interference phases to perform atmospheric phase correction; performing phase filtering processing on the interference phases after the atmospheric phase correction; performing phase unwrapping on the interference phases after the filtering processing; and performing geocoding processing to complete deformation extraction. Through near-synchronous imaging of multiple platforms on the same region, the generation path of time dimension interference is cut off from the physical layer, the influence of short-term atmospheric fluctuation on the phase is greatly weakened, noise and ambiguity in the phase unwrapping process are further removed in combination with the redundancy of multi-view observation of the distributed system, and high-precision extraction of the surface deformation field is realized.

Description

Technical Field

[0001] This invention belongs to the field of satellite image processing, and in particular relates to a method for atmospheric phase correction and deformation extraction in a distributed InSAR system. Background Technology

[0002] Synthetic Aperture Radar Interferometry (InSAR) technology, with its significant advantages of all-day, all-weather, and large-scale monitoring, has become one of the core methods for monitoring surface deformation. It has been widely used in scenarios such as monitoring crustal movement after earthquakes, monitoring ground subsidence caused by groundwater extraction in urban areas, and detecting minute deformations in volcanic activity areas.

[0003] However, traditional InSAR technology faces numerous challenges in practical applications. Atmospheric phase delay is the most prominent issue. As a non-homogeneous medium, the refractive index of the atmosphere varies complexly across time and space due to factors such as water vapor, temperature, and air pressure, causing the radar signal propagation path to bend and introducing phase errors into the interferogram. The interference signal generated by this atmospheric phase delay is similar in magnitude to the actual surface deformation signal. In some extreme weather conditions, such as heavy rain or severe convective weather, the intensity of the atmospheric phase delay signal far exceeds the deformation signal, completely masking the deformation signal and severely reducing the accuracy and reliability of deformation extraction results. Existing methods for modeling and estimating atmospheric phase delay using external atmospheric data (such as meteorological numerical models, GPS data, and imaging spectrometer data) have some corrective effect on large and medium-scale atmospheric turbulence effects. However, when facing complex and variable small-scale atmospheric turbulence, the correction effect is significantly reduced due to limitations in spatiotemporal resolution and weather interference.

[0004] Furthermore, atmospheric correction methods based on InSAR data itself, such as stacking InSAR and interferogram overlay atmospheric estimation methods, suffer from insufficient applicability when dealing with nonlinear deformation regions caused by urban construction and mining, due to the assumption of linear surface deformation. While PS or DS methods can suppress atmospheric effects to some extent through phase difference between adjacent points, they are insufficient to effectively remove atmospheric interference when small-scale atmospheric effects occur because of significant differences in atmospheric characteristics between closely spaced areas. In vegetated areas, dense vegetation causes multiple scattering and absorption of radar signals, leading to severe surface decoherence. Strong scattering point targets (such as buildings and rock outcrops), upon which traditional InSAR technology relies, are extremely rare in high-vegetation-coverage areas, significantly reducing the density of monitoring points and making it impossible to obtain sufficient deformation information. Summary of the Invention

[0005] This invention aims to solve the above problems by providing a method for atmospheric phase correction and deformation extraction in a distributed InSAR system.

[0006] In a first aspect, the present invention provides a method for atmospheric phase correction and deformation extraction of a distributed InSAR system, comprising: acquiring near-synchronous observation data of multiple satellites in a distributed InSAR system and performing image registration;

[0007] The interferometric phase is obtained by performing conjugate multiplication on the registered images;

[0008] The atmospheric phase is corrected by subtracting the phases of the two interference frames.

[0009] Phase filtering is performed on the interferometric phase after atmospheric phase correction;

[0010] Phase unwrapping is performed on the filtered interference phase;

[0011] Geocoding is performed to convert all products to a map projection coordinate system, and deformation extraction is completed.

[0012] The acquired near-synchronous imaging data ensures that time-related interference in the raw data is minimized, reducing initial noise for subsequent processing. Based on the phase difference of the synchronous data, the spatial distribution characteristics of the atmospheric phase are directly separated without relying on external models, achieving accurate removal of atmospheric interference and keeping the residual atmospheric phase error within millimeters. Finally, combined with the redundancy of multi-view observations in the distributed system (the response of deformation phase under different views can be cross-verified), noise and ambiguity in the phase unwrapping process are further eliminated, ultimately achieving high-precision extraction of the surface deformation field.

[0013] Furthermore, the atmospheric phase correction and deformation extraction method for the distributed InSAR system described in this invention includes image registration consisting of sequential coarse registration at the pixel level and fine registration at the sub-pixel level. The main difference between the two registration steps is their precision; the offset in coarse registration is generally at the pixel level (greater than 1), while the offset in fine registration is generally at the sub-pixel level (less than 1).

[0014] Furthermore, in the distributed InSAR system atmospheric phase correction and deformation extraction method of the present invention, the step of obtaining the interferometric phase by performing conjugate multiplication on the registered images specifically includes:

[0015] A distributed InSAR system uses multiple satellites to independently detect the same ground target twice, acquiring two sets of echo signals corresponding to the target; the echo signals are represented by complex numbers as follows:

[0016] S = a + b · i = A ·exp[ i · ψ ];

[0017] In the formula, a , b These are the real and imaginary parts of a complex number, respectively. i Represents the imaginary unit; A Indicates the echo amplitude; ψ Indicates the phase of the echo;

[0018] The phase of the echo signal is represented as follows:

[0019] ;

[0020] In the formula, arg[] represents the argument operation; U The additional phase contribution representing the ground target; R The slant range from the radar antenna to the target point; l The radar wavelength;

[0021] The phase difference between the two echo signals is expressed as:

[0022] ;

[0023] In the formula, R 1 is the slant distance of the main star. R 2 represents the slant distance of the auxiliary star;

[0024] Phase interference is achieved by conjugate multiplication of the corresponding pixel values ​​affected by pairing:

[0025] INT = S 1· S 2 * = A 1· A 2·exp[ i ·( ψ 1 -ψ 2)

[0026] In the formula, INT The complex number of interference is obtained by multiplying the complex echoes of the main image and the auxiliary image at the same pixel by their conjugates. S 1 The main image complex signal S 2 For auxiliary image complex signals, * Indicates complex conjugation; A 1 , A 2 Primary and secondary echo amplitudes, ψ 1, ψ 2 is the main and auxiliary echo phase;

[0027] The interference phase is achieved by taking the argument of the above equation:

[0028] Ø =arg[ INT ];

[0029] In the formula, Ø The interference phase is represented by the complex number of interference phases. INT Argument: arg[] represents the phase angle when taking a complex number.

[0030] Furthermore, in the atmospheric phase correction and deformation extraction method for the distributed InSAR system described in this invention, the phase filtering employs mean filtering. Even after atmospheric phase correction, noise interference still exists in the interferometric phase image, manifested as numerous cluttered points on the image. These noise points are particularly chaotic, leading to discontinuities and inconsistencies in the phase data. This is highly detrimental to subsequent unwrapping and easily results in erroneous conclusions. To improve accuracy, filtering of the interferometric phase image is necessary.

[0031] Furthermore, the atmospheric phase correction and deformation extraction method for the distributed InSAR system described in this invention includes the following geocoding process: First, by combining the satellite orbital state vector and the SAR imaging geometric model, the three-dimensional geodetic coordinates of the ground point corresponding to each pixel in the SAR image are determined through spatial geometric calculations; then, the three-dimensional geodetic coordinates are transformed into a two-dimensional map plane coordinate system through map projection to generate a surface deformation field in the form of an irregular grid; finally, the irregular grid data is processed through an interpolation algorithm to obtain a spatially uniform regular grid deformation result.

[0032] Furthermore, the atmospheric phase correction and deformation extraction method for the distributed InSAR system described in this invention includes Gaussian projection and transverse Mercator projection as the map projection method.

[0033] In a second aspect, the present invention provides a distributed InSAR system atmospheric phase correction and deformation extraction system, including an image acquisition module, a phase correction module, a phase filtering and unwrapping module, and a deformation extraction output module.

[0034] The image acquisition module is used to acquire near-synchronous observation data from multiple satellites of the distributed InSAR system and perform image registration.

[0035] The phase correction module is used to perform conjugate multiplication on the registered images to obtain the interferometric phase; and to perform atmospheric phase correction by subtracting the two interferometric phases.

[0036] The phase filtering and unwrapping module is used to perform phase filtering on the atmospheric phase-corrected interference phase and to perform phase unwrapping on the filtered interference phase.

[0037] The deformation extraction and output module is used for geocoding, converting all products to a map projection coordinate system, and outputting the results after deformation extraction.

[0038] Furthermore, in the distributed InSAR system atmospheric phase correction and deformation extraction system of the present invention, when the deformation extraction output module performs geocoding processing, it first combines the satellite orbital state vector and the SAR imaging geometric model to determine the three-dimensional geodetic coordinates of the ground point corresponding to each pixel in the SAR image through spatial geometric calculation; then, it transforms the three-dimensional geodetic coordinates to a two-dimensional map plane coordinate system through map projection to generate a surface deformation field in the form of an irregular grid; finally, it processes the irregular grid data through an interpolation algorithm to obtain a spatially uniform regular grid deformation result.

[0039] Thirdly, the present invention provides an atmospheric phase correction and deformation extraction device for a distributed InSAR system, comprising a memory and a processor; the memory is used to store a computer program; the processor is used to implement the atmospheric phase correction and deformation extraction method for a distributed InSAR system as described in any one of the first aspects when the computer program is executed.

[0040] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the atmospheric phase correction and deformation extraction method for a distributed InSAR system as described in any one of the first aspects.

[0041] The atmospheric phase correction and deformation extraction method for distributed InSAR systems described in this invention relies on the multi-satellite formation collaborative observation architecture of a spaceborne distributed InSAR system. Through near-synchronous Earth imaging of the same area by multiple platforms (with time difference approaching zero), it physically cuts off the path of temporal interference, significantly reducing the impact of short-term atmospheric fluctuations on the phase. This provides a "low-interference, high-consistency" raw data foundation for subsequent data processing, overcoming the limitations of traditional InSAR's "introducing interference first and then passively correcting" approach. By performing phase difference analysis on multiple synchronously acquired data sets, the spatial distribution characteristics of the atmospheric phase can be directly separated without relying on external models. Quantitative atmospheric phase stripping can be completed solely through the inherent correlation of the system's own observation data, improving the universality of atmospheric correction and avoiding the transmission of external data errors to the atmospheric correction results. Combined with the redundancy of multi-view observations in the distributed system, noise and ambiguity in the phase unwrapping process are further eliminated, ultimately achieving high-precision extraction of the surface deformation field. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the atmospheric phase correction and deformation extraction method for the distributed InSAR system according to an embodiment of the present invention;

[0043] Figure 2 This is a schematic diagram of the interference phase described in an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of repeated SAR observations as described in an embodiment of the present invention;

[0045] Figure 4 This is a schematic diagram of the untangling result described in an embodiment of the present invention;

[0046] Figure 5 This is a schematic diagram of the deformation result after geocoding as described in an embodiment of the present invention. Detailed Implementation

[0047] The atmospheric phase correction and deformation extraction method for the distributed InSAR system described in this invention will be explained in detail below with reference to the accompanying drawings and embodiments.

[0048] This embodiment discloses a method for atmospheric phase correction and deformation extraction in a distributed InSAR system, such as... Figure 1 As shown, the specific process includes the following:

[0049] Acquire near-synchronous observation data from multiple satellites of the distributed InSAR system and perform image registration;

[0050] The interferometric phase is obtained by performing conjugate multiplication on the registered images;

[0051] The atmospheric phase is corrected by subtracting the phases of the two interference frames.

[0052] Phase filtering is performed on the interferometric phase after atmospheric phase correction;

[0053] Phase unwrapping is performed on the filtered interference phase;

[0054] Geocoding is performed to convert all products to a map projection coordinate system, and deformation extraction is completed.

[0055] The spaceborne distributed InSAR system breaks through the traditional design framework of single-satellite InSAR. It is a novel satellite system concept built by organically combining high-precision deformation monitoring technology of InSAR with multi-platform collaborative observation technology of satellite formation. The spaceborne distributed InSAR system consists of multiple SAR satellites working collaboratively, and can flexibly switch between multiple operating modes such as single-transmission single-reception, single-transmission dual-reception, and dual-transmission dual-reception. It not only possesses significant advantages such as wide Earth coverage, short observation revisit cycle, and low system construction and maintenance costs, but also demonstrates outstanding potential in the field of atmospheric phase correction. The following embodiments are verified using measured data from the TH-2 satellite, specifically including:

[0056] Because of slight differences in the radar's trajectories during the two flights, the pixels in the two images do not correspond perfectly, exhibiting pixel offsets in both the azimuth and range directions. Therefore, to ensure that pixels at the same location in the primary and secondary images correspond to the same target point on the ground, image registration is required. This image registration includes sequential coarse registration at the pixel level and fine registration at the sub-pixel level. The main difference between the two registration steps is their precision: coarse registration typically involves offsets at the pixel level (greater than 1), while fine registration typically involves offsets at the sub-pixel level (less than 1).

[0057] In this embodiment of the disclosure, within the InSAR observation system, the radar antenna performs two independent detections of the same ground target. By receiving the electromagnetic waves reflected after the radar waves illuminate the target, two sets of echo signals corresponding to the target are ultimately obtained. According to the wave equation, the echo signal is represented by a complex number as:

[0058] S = a + b · i = A ·exp[ i · ψ ];

[0059] In the formula, a , b These are the real and imaginary parts of a complex number, respectively. i Represents the imaginary unit; A Indicates the echo amplitude; ψ Indicates the phase of the echo.

[0060] By combining the geometric principles of SAR imaging with the analysis of the reflection characteristics of radar waves by ground resolution units, it can be seen that the phase information of the radar echo signal... ψ It comprises two core components: the first is the path-dependent phase, determined by the slant range (i.e., propagation path length) of the radar wave from the antenna to the ground target point, directly related to the propagation distance of the electromagnetic wave; the second is the scattering-added phase, originating from the scattering of radar waves by various scatterers within the ground resolution cell (such as surface undulations and ground structures), and is a key phase component reflecting the scattering characteristics within the resolution cell. Therefore, the echo signal phase is represented as:

[0061] ;

[0062] In the formula, arg[] represents the argument operation; U The additional phase contribution representing the ground target; R The slant range from the radar antenna to the target point; l This is the radar wavelength.

[0063] If the ground scattering characteristics remain stable during the interval between two radar observations of the target (i.e., there is no time decorrelation due to changes in surface cover, movement of scattering objects, etc.), then the random scattering phase generated by ground scattering in the radar echo signals acquired from the two observations will be consistent. The phase difference between the two echo signals is then expressed as:

[0064] ;

[0065] In the formula, R 1 is the slant distance of the main star. R 2 represents the slant distance of the auxiliary star.

[0066] Therefore, the phase difference generated by the two imaging processes is strictly proportional to the change in the radar wave propagation path. In the InSAR interferometric modeling and solution process, phase interferometry is achieved by multiplying the corresponding pixel values ​​of the paired effects using conjugate multiplication.

[0067] INT = S 1· S 2 * = A 1· A 2·exp[ i ·( ψ 1 -ψ 2)

[0068] In the formula, INT The complex number of the interference (the complex representation of the interference phase) is obtained by multiplying the complex echoes of the main image and the auxiliary image at the same pixel by conjugate; S 1 The main image complex signal S 2 For auxiliary image complex signals, * Indicates complex conjugation; A 1 , A 2 Primary and secondary echo amplitudes, ψ 1, ψ 2 represents the primary and secondary echo phases.

[0069] The interference phase is achieved by taking the argument of the above equation:

[0070] Ø =arg[ INT ];

[0071] In the formula, Ø The interference phase is represented by the complex number of interference phases. INT Argument (phase angle): arg[] represents the phase angle taken as a complex number.

[0072] The above equation shows that the range of values ​​for the interference phase is [- π , π ] is a phase value that is less than a whole period, referred to as the principal phase value or entanglement value. In this embodiment, such as Figure 2 The image shows the generated interferometric phase map (interferogram), which displays the entangled interferometric phase distribution of pixels within the study area in pseudo-color mode. The colors correspond to the principal phase value intervals [- π , π The continuous colored bands (interference fringes) in the figure represent phase contour lines. The denser the fringes, the faster the phase change, which is usually related to factors such as topographic relief, deformation, or atmospheric disturbance.

[0073] In this embodiment of the disclosure, two radar images (including image 1 and image 3) taken at the same time before deformation occurs and two radar images (including image 2 and image 4) taken at the same time after deformation occur are acquired by a spaceborne distributed InSAR system. Interference is performed using image 1 and image 3 and image 2 and image 4 respectively. Both images contain terrain phase, atmospheric phase and deformation phase. However, since the deformation components along the line of sight are different, high-precision deformation information can be calculated.

[0074] Through such Figure 3 As shown in the schematic diagram of repeated SAR observations, the figure... R 2 、 H represents the slant distance from the secondary star to the ground point, and H represents the altitude of the primary star. a Satellite baseline B Angle with the horizontal direction , The projection of the deformation perpendicular to the ground surface onto the radar line of sight is related to the radar's downward angle of view. Therefore, due to the change in the downward angle of view, the deformation component along the line of sight will also change. The relationship between the two is as follows:

[0075] ;

[0076] In the formula, Δ H The deformation perpendicular to the Earth's surface. i For radar downward view, Δ r This represents the deformation component along the line of sight.

[0077] The signal information of the target points obtained from Images 1 and 2 is represented as follows:

[0078] S 1 ( R 1) =| S 1 ( R 1) |exp() i · ψ 1);

[0079] S 2 (R 2) =| S 2 ( R 2) |exp() i · ψ 2);

[0080] In the formula ，R 1 is the slant distance of the main star. R 2 represents the slant distance of the auxiliary star; i Represents the imaginary unit. S 1 (R 1 ) Main echo amplitude, S 2 (R 2 ) To supplement the echo amplitude.

[0081] The signal phases acquired by satellite S1 and satellite S2 can be obtained from the above formula. Represented as:

[0082] ;

[0083] ;

[0084] By interferometric processing of images 1 and 2, the interferometric phase containing surface displacement can be obtained. ψ 12 value:

[0085] ;

[0086] Since the deformation is much smaller than the slant range of the auxiliary star, the interference phase... ψ 12 Further expressed as:

[0087] ;

[0088] The first term in the above interferometric phase is the interferometric phase of the target point before deformation. Since no atmospheric phase correction has been performed, the interferometric phase at this time includes the reference surface phase, the terrain phase, and the atmospheric phase. Therefore, after the ground deformation, the interferometric phase of the target point is:

[0089] ;

[0090] ;

[0091] In the formula, For the reference plane phase, For terrain phase, Atmospheric phase, The vertical baseline between satellite S1 and satellite S2, The parallel baseline between satellite S1 and satellite S2, lFor radar wavelength, i 1 represents the radar downward view of satellite S1. h This represents the elevation of a ground point.

[0092] Therefore, it is evident that the interferometric phase includes the atmospheric phase. Conventional atmospheric correction methods cannot completely eliminate the influence of the atmosphere on the deformation results. The deformation phase also contains a small amount of atmospheric phase. To address this issue, in this embodiment, distributed InSAR is used to acquire different images of the same observation area at the same time, thereby ensuring that their interferometric phases contain the same atmospheric component. Using the same calculation steps as described above, the same interferometric processing is performed on images 3 and 4, i.e., the interferometric phase... ψ 34 for:

[0093] ;

[0094] ;

[0095] ;

[0096] In the formula, The vertical baseline between satellites S3 and S4. The parallel baseline between satellites S3 and S4 l For radar wavelength, i 3 represents the radar downward view of satellite S3. R 3 represents the slant range of satellite S3. R 4 represents the slant range of satellite S4; h This represents the elevation of a ground point.

[0097] After interferometric processing of images 1 and 2, and images 3 and 4 respectively, it is obvious that the interferometric phases all contain atmospheric phase. Since the distributed InSAR system acquires images at the same time, the atmospheric phase components contained in their interferometric phases are consistent. Therefore, subtracting the two interferometric phases can achieve a high-precision atmospheric phase correction. The specific calculation steps are as follows:

[0098] ;

[0099] ;

[0100] ;

[0101] ;

[0102] The above steps can achieve high-precision correction of atmospheric phase, thereby calculating the deformation perpendicular to the Earth's surface.

[0103] In this embodiment, even after atmospheric phase correction, noise interference still exists in the interferometric phase image, manifested as numerous messy points on the image. These noise points are particularly chaotic, leading to discontinuities and inconsistencies in the phase data. This is highly detrimental to subsequent unwrapping and easily results in erroneous conclusions. To improve accuracy, the interferometric phase image must be filtered.

[0104] In this embodiment, the mean filtering method is specifically used. Let matrix s be... i Complex data is stored in a specific format, and the filtered data is stored in matrix s. f In the middle, a 9x9 sliding window filter is used, s f The calculation is performed using the following formula:

[0105] ;

[0106] In the formula, s i This represents a matrix storing complex data, where m and n represent pixel coordinate indices (row and column), and i and j represent relative offsets within the sliding window (offset i in the row direction and offset j in the column direction).

[0107] The phase in the interferometric phase diagram is a wrapped value located in the interval (-π, π]. This wrapped value cannot be directly converted into deformation, making it impossible to directly obtain the true deformation information. Therefore, the interferometric phase needs to be unwrapped. There is a phase difference between the true phase and the wrapped phase that is an integer multiple of 2π. Without constraints, the true value will have multiple solutions. Therefore, all unwrapping algorithms must follow certain assumptions, namely, that in most areas of the interferometric diagram, due to a sufficiently high spatial sampling rate, the true phase difference between any two adjacent pixels is less than 2π. Currently, there are two main research methods for phase unwrapping: one is the path integral method, and the other is an algorithm based on the minimum norm. In this embodiment, different strategies are used to select the integration path to avoid discontinuities in the data, thereby achieving the condition of phase gradient closure and zero path integral. In this embodiment, Figure 4 This is a schematic diagram of the unfolded phase distribution after phase unwrapping of the interferometric phase map. The phase distribution differs from the unwrapped phase distribution only in the [-] section. π , π The internal cycle changes differently; after untangling, the phase changes continuously in space. Originally, it was 2... π The phase jumps caused by the periodicity are eliminated, making the overall phase field smoother and more coherent. Different colors in the figure correspond to different unfolded phase magnitudes, and regions with larger phase gradients reflect more drastic phase changes in those regions, providing usable continuous phase input for deformation inversion.

[0108] In this embodiment, the geocoding process is based on a reference ellipsoid frame (such as the WGS84 ellipsoid) that matches the satellite orbital coordinate system: First, combining the satellite orbital state vector (including parameters such as position and velocity) with the SAR imaging geometric model (such as the range-Doppler model), spatial geometric calculations are used to determine the three-dimensional geodetic coordinates (geodetic longitude, geodetic latitude, and geodetic height) of the ground point corresponding to each pixel in the SAR image; then, any map projection method, such as Gaussian projection or transverse Mercator projection, is selected to transform the three-dimensional geodetic coordinates to a two-dimensional map plane coordinate system, generating a non-regular grid-form surface deformation field; finally, the non-regular grid data is processed using interpolation algorithms (such as Kriging interpolation and bilinear interpolation) to obtain a spatially uniform regular grid deformation result. In this embodiment, Figure 4 The deformation result after geocoding of the unwrapping result is shown below. Figure 5 As shown, Figure 5 The deformation magnitude is represented by color gradients, with different colors corresponding to different amplitude ranges. The overall deformation exhibits a clear non-uniform distribution in space. This result eliminates the influence of imaging geometry while achieving a consistent expression of deformation information and geographic coordinates, providing a unified spatial reference for subsequent deformation time series analysis and visualization, thus meeting the needs of subsequent deformation analysis and visualization.

[0109] This embodiment leverages the efficient observation capabilities of spaceborne distributed InSAR near-synchronous imaging and the rapid processing advantages of data-dependent atmospheric correction to achieve integrated collaboration of "efficient observation and precise processing": On the one hand, near-synchronous observation by multi-satellite formations can significantly shorten the data acquisition cycle, meeting the timeliness requirements of short-cycle deformation monitoring. It can adapt to conventional scenarios such as long-term urban subsidence monitoring, as well as special scenarios such as emergency response to natural disasters, further expanding the application boundaries of InSAR technology in the field of high-precision deformation monitoring.

[0110] This second embodiment discloses a distributed InSAR system atmospheric phase correction and deformation extraction system, including an image acquisition module, a phase correction module, a phase filtering and unwrapping module, and a deformation extraction output module;

[0111] The image acquisition module is used to acquire near-synchronous observation data from multiple satellites of the distributed InSAR system and perform image registration.

[0112] The phase correction module is used to perform conjugate multiplication on the registered images to obtain the interferometric phase; and to perform atmospheric phase correction by subtracting the two interferometric phases.

[0113] The phase filtering and unwrapping module is used to perform phase filtering on the atmospheric phase-corrected interference phase and to perform phase unwrapping on the filtered interference phase.

[0114] The deformation extraction and output module is used for geocoding processing, converting all products to a map projection coordinate system and outputting the extracted deformation. During geocoding, the system first combines satellite orbital state vectors with the SAR imaging geometric model to determine the three-dimensional geodetic coordinates of the ground point corresponding to each pixel in the SAR image through spatial geometric calculations. Then, the three-dimensional geodetic coordinates are converted to a two-dimensional map plane coordinate system through map projection, generating a non-irregular grid-like surface deformation field. Finally, the non-irregular grid data is processed using an interpolation algorithm to obtain a spatially uniform regular grid deformation result.

[0115] The specific operation steps of the distributed InSAR system atmospheric phase correction and deformation extraction system described in this embodiment are the same as those of the distributed InSAR system atmospheric phase correction and deformation extraction method described in Embodiment 1 above, and will not be repeated here.

[0116] This embodiment three discloses a distributed InSAR system atmospheric phase correction and deformation extraction device, including a memory and a processor; the memory is used to store a computer program; the processor is used to implement the distributed InSAR system atmospheric phase correction and deformation extraction method described in embodiment one when the computer program is executed. The specific extraction method steps are the same as those in the aforementioned embodiment one, and will not be repeated here.

[0117] This embodiment four discloses a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the atmospheric phase correction and deformation extraction method for the distributed InSAR system described in embodiment one. The specific extraction method steps are the same as those in embodiment one, and will not be repeated here.

[0118] The computer described in this application embodiment can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. The computer-readable storage medium can be any usable medium that a computer can read, or a data storage device such as a server or data center that integrates one or more usable media. The usable medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital versatile optical disc (DVD)), or a semiconductor medium (e.g., solid-state drive (SSD)). The software formed by the computer's stored code can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other storage media that are mature in the art.

[0119] In the various embodiments of this application, the functional modules can be integrated into one processing unit or module, or each module can exist physically separately, or two or more modules can be integrated into one unit or module. In the above embodiments, they can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, they can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated.

[0120] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for atmospheric phase correction and deformation extraction based on a distributed InSAR system, characterized in that... include: Acquire near-synchronous observation data from multiple satellites of a distributed InSAR system. The near-synchronous observation data includes at least two sets of image pairs corresponding to the period before and after deformation, respectively. Perform image registration. The registered images are multiplied by conjugate to obtain the interference phase; the interference phase includes two interference phases obtained by interferometry from image pairs at the same time before deformation and image pairs at the same time after deformation. The atmospheric phase is corrected by subtracting the phases of the two interference frames. Phase filtering is performed on the interferometric phase after atmospheric phase correction; Phase unwrapping is performed on the filtered interference phase; Geocoding is performed to convert all products to a map projection coordinate system, and deformation extraction is completed.

2. The atmospheric phase correction and deformation extraction method based on a distributed InSAR system according to claim 1, characterized in that: The image registration includes sequential coarse registration at the pixel level and fine registration at the sub-pixel level.

3. The atmospheric phase correction and deformation extraction method based on a distributed InSAR system according to claim 1, characterized in that, The specific steps of obtaining the interference phase by performing conjugate multiplication on the registered images include: A distributed InSAR system uses multiple satellites to independently detect the same ground target twice, acquiring two sets of echo signals corresponding to the target; the echo signals are represented by complex numbers as follows: S = a + b·i = A·exp[i·ψ]; In the formula, a and b are the real and imaginary parts of the complex number, respectively; i represents the imaginary unit; A represents the echo amplitude; and ψ represents the phase of the echo. The phase of the echo signal is represented as follows: ； In the formula, arg[] represents the argument operation; U represents the additional phase contribution of the ground target; R is the slant range from the radar antenna to the target point; λ is the radar wavelength; The phase difference between the two echo signals is expressed as: ； In the formula, R1 is the slant range of the primary star and R2 is the slant range of the secondary star. Phase interference is achieved by conjugate multiplication of the corresponding pixel values ​​affected by pairing: INT=S1·S2 * =A1·A2·exp[i·(ψ1-ψ2)]; In the formula, INT represents the complex interference number, which is obtained by multiplying the complex echoes of the main image and the auxiliary image at the same pixel by conjugate; S1 is the complex signal of the main image, S2 is the complex signal of the auxiliary image, * represents complex conjugate; A1 and A2 are the amplitudes of the main and auxiliary echoes, and ψ1 and ψ2 are the phases of the main and auxiliary echoes. The interference phase is achieved by taking the argument of the above equation: Ø = arg[INT]; In the formula, Ø represents the interference phase, that is, the argument of the complex interference number INT: arg[ ] represents the phase angle of the complex number.

4. The atmospheric phase correction and deformation extraction method based on a distributed InSAR system according to claim 1, characterized in that: The phase filtering uses mean filtering.

5. The atmospheric phase correction and deformation extraction method based on a distributed InSAR system according to claim 1, characterized in that, The geocoding process includes: first, combining the satellite orbital state vector with the SAR imaging geometric model, and determining the three-dimensional geodetic coordinates of the ground point corresponding to each pixel in the SAR image through spatial geometric calculations; then, transforming the three-dimensional geodetic coordinates to a two-dimensional map plane coordinate system through map projection, generating a surface deformation field in the form of an irregular grid; finally, processing the irregular grid data through an interpolation algorithm to obtain a spatially uniform regular grid deformation result.

6. The atmospheric phase correction and deformation extraction method based on a distributed InSAR system according to claim 5, characterized in that; The map projection methods include Gaussian projection and transverse Mercator projection.

7. An atmospheric phase correction and deformation extraction system based on a distributed InSAR system, characterized in that: It includes an image acquisition module, a phase correction module, a phase filtering and unwrapping module, and a deformation extraction and output module; The image acquisition module is used to acquire near-synchronous observation data from multiple satellites of the distributed InSAR system. The near-synchronous observation data includes at least two sets of image pairs corresponding to the period before and after deformation, respectively, for image registration. The phase correction module is used to perform conjugate multiplication on the registered images to obtain the interference phase; the interference phase includes two interference phases obtained by interferometry from image pairs at the same time before deformation and image pairs at the same time after deformation; and atmospheric phase correction is performed by subtracting the two interference phases. The phase filtering and unwrapping module is used to perform phase filtering processing on the interference phase after atmospheric phase correction; And perform phase unwrapping on the filtered interference phase; The deformation extraction and output module is used for geocoding, converting all products to a map projection coordinate system, and outputting the results after deformation extraction.

8. The atmospheric phase correction and deformation extraction system based on a distributed InSAR system according to claim 7, characterized in that: When performing geocoding processing, the deformation extraction output module first combines the satellite orbital state vector with the SAR imaging geometric model to determine the three-dimensional geodetic coordinates of the ground point corresponding to each pixel in the SAR image through spatial geometric calculations. Then, it transforms the three-dimensional geodetic coordinates to a two-dimensional map plane coordinate system through map projection to generate a surface deformation field in the form of an irregular grid. Finally, it processes the irregular grid data through an interpolation algorithm to obtain a regular grid deformation result with uniform spatial distribution.

9. An atmospheric phase correction and deformation extraction device based on a distributed InSAR system, characterized in that: It includes a memory and a processor; the memory is used to store a computer program; the processor is used to implement the atmospheric phase correction and deformation extraction method based on a distributed InSAR system as described in any one of claims 1-6 when the computer program is executed.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the atmospheric phase correction and deformation extraction method based on a distributed InSAR system as described in any one of claims 1-6.

Images