A method for correcting the atmospheric interference phase of a ground-based circular-arc interferometric synthetic aperture radar

Abstract

The present invention is applicable to the field of surveying and mapping science and technology, and relates to a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar, including: constructing a differential interferometric phase model of the ground-based circular arc interferometric synthetic aperture radar; eliminating phase noise; screening permanent scatterers; dividing the obtained interferometric phase in the azimuth direction into subspaces, then dividing the subspace grids according to a set window size, and performing phase weighted average processing on the permanent scatterers in the grids according to the correlation coefficient and amplitude deviation, and taking it as the atmospheric interference phase sample point of the grid center point; screening the obtained atmospheric interference phase sample points to obtain available sample points; using the available sample points to estimate the parameters of the atmospheric interference phase error model, and further obtaining the atmospheric interference phase error of the detection area; and obtaining the deformation phase. The method of the present invention is simple and efficient, does not require the introduction of external auxiliary data, but only relies on the interferometric phase to correct the atmospheric interference phase, and has a high correction accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of surveying and mapping science and technology, and particularly relates to a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar. Background Art

[0002] SAR (Synthetic Aperture Radar), as an all-weather, all-day, and high-resolution imaging radar, is widely used in the field of remote sensing. Since SAR images have phase information, non-contact and high-precision extraction of deformation information in the imaging area can be achieved through interferometric processing of SAR image sequences. Given the above advantages of the SAR system, it has become one of the important means for ground deformation monitoring. Traditional SARs are mostly carried on platforms such as satellites and airplanes, which makes the revisit time of the radar very long and unable to continuously monitor specific targets. Ground-Based SAR (GBSAR) is a miniaturized and low-cost SAR system that can achieve continuous observation of a specific area. It has a short revisit period and high monitoring accuracy. Combining interferometry and differential interferometry techniques, it can continuously monitor the deformation of a specific scene for a long time and is widely used in disaster warnings such as collapses and landslides. However, usually GBSARs are installed on straight tracks. Due to the limited track length, the observation range of GBSAR in the azimuth direction is restricted. Another type, the Rotor SAR (ROSAR), proposed in the early 1990s, obtains a wider observation range by placing the radar on a rotating robotic arm. ROSAR is usually installed on a helicopter rotor and uses a pulsed signal, which makes the ROSAR system very complex. Based on the motion form of ROSAR, an improved rotating SAR using the Frequency Modulated Continuous Wave (FMCW) system was proposed in 2012 and named Arc-SAR. Arc-InSAR is a high-precision deformation measurement technique that combines interferometry with Arc-SAR. It performs differential interferometric processing on two SAR images obtained at the same location at different times and realizes deformation measurement based on phase information, enabling continuous monitoring of a large area and high precision of the ground within a limited space. Arc-InSAR generally operates in the X or Ku band, and the deformation measurement accuracy can reach the millimeter or sub-millimeter level.

[0003] Atmospheric interference is the phase error caused by different phase delays due to factors such as atmospheric pressure, temperature, and humidity that change with time and space during the propagation of radar electromagnetic waves in the air. Atmospheric interference seriously affects the deformation inversion accuracy and reliability of InSAR interferometry, and is one of the important limiting factors hindering the application development of InSAR interferometry. For Arc-InSAR with the ability to measure large-scale deformations, atmospheric interference is an actual problem that needs to be urgently solved. Currently, InSAR atmospheric interference phase correction methods are mainly divided into two categories: (1) based on external auxiliary data; (2) based on the interference phase itself. Methods based on external auxiliary data mainly include modeling based on surface meteorological data, retrieving atmospheric interference phase based on GPS data, and measurement methods based on space radiometers. These phase correction methods based on external auxiliary data are mainly affected by the low spatial resolution of the auxiliary data, and the correction effect is often not very satisfactory. Correction methods based on the interference phase itself include interference phase filtering methods, correction methods based on time series SAR images, and atmospheric interference phase correction algorithms based on stable control points. Among them, interference phase filtering methods and correction methods based on time series SAR images also face the problem of separating atmospheric interference phase from non-linear deformations. Therefore, the most widely used InSAR atmospheric interference phase correction method currently is the atmospheric interference phase correction based on stable control points. Although this method has a good correction effect, it is proposed under the condition of assuming that the atmospheric interference phase delay is consistent in the azimuth direction, and does not consider the spatial variation characteristics of the atmospheric interference phase in the azimuth direction in complex situations, which will cause the problem of under-correction in the azimuth direction during the correction process of spatially variable atmospheric interference phase.

[0004] Therefore, how to propose a method that can correct the atmospheric interference phase without introducing external auxiliary data and only relying on the interference phase itself is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar, so as to solve the problems in the prior art that the method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar needs to introduce external data and has low accuracy.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The present invention provides a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar, including the following steps:

[0008] S10. Construct a differential interference phase model of a ground-based circular arc interferometric synthetic aperture radar, and the phase model can be specifically expressed as:

[0009]

[0010] Among them, is the interference phase of the main image u and the slave image v, is the deformation phase, is the atmospheric interference phase, is the noise phase;

[0011] S20. Filter the interference phase to eliminate phase noise;

[0012] S30. Screen out permanent scatterers from the interference phase according to the statistical characteristic quantity, where the permanent scatterer is the control point where the interference phase is approximately zero;

[0013] S40. Perform subspace partitioning on the interference phase obtained in step S20 in the azimuth direction, then divide the subspace into grids according to the set window size, and perform phase weighted average processing on the permanent scatterers in the grid according to the correlation coefficient and amplitude deviation, and use it as the atmospheric interference phase sample point of the grid center point;

[0014] S50. Screen the atmospheric interference phase sample points in each subspace obtained in step S40 to obtain available sample points;

[0015] S60. Use the available sample points to estimate the atmospheric interference phase error model parameters of the corresponding subspace, and then obtain the atmospheric interference phase error of the corresponding detection subspace;

[0016] S70: Obtain the compensated deformation phase of all detection spaces.

[0017] Furthermore, the filtering process in step S20 is weighted circular median filtering, and the weighting coefficient is:

[0018]

[0019] Among them, (kn, km) is the coordinate serial number of the permanent scatterer in the interference image, n is the abscissa of the interference phase to be filtered, and m is the ordinate of the interference phase to be filtered, Among them, M is the length of the filtering window on the horizontal axis, and N is the length of the filtering window on the vertical axis,

[0020] Furthermore, the steps of screening out permanent scatterers in step S30 include:

[0021] S31. Calculate the correlation coefficient γ according to the main and slave complex scattering images, and the formula is as follows:

[0022]

[0023] Among them, M(i, j) and the said S(i, j) are the main complex scattering image and the slave complex scattering image that constitute the interference phase respectively, * represents complex conjugate, w1 is the size of the first sliding window, and w2 is the size of the second sliding window;

[0024] S32. Using N p frames of temporal complex scattering images, the average correlation coefficient can be calculated The formula is as follows:

[0025]

[0026] Among them, q represents the q-th frame of the interference phase, and γ q represents the correlation coefficient of the q-th frame;

[0027] S33. Set the first threshold T 1 , and screen out the pixel points as the candidate points of the permanent scatterers;

[0028] S34. Calculate the amplitude deviation D A of the temporal complex scattering image, and the formula is as follows:

[0029]

[0030] Among them, δ A represents the standard deviation of the pixel point amplitude time series, and m A is the mean value of the pixel point amplitude time series;

[0031] S35. Set the second threshold T 2 , and screen out the D A ≤T 2 permanent scatterers from the candidate point set of the permanent scatterers obtained in step S31.

[0032] Furthermore, the first threshold T 1 in step S32 is set to 0.9, and the second threshold T 2 is set to 0.1.

[0033] Furthermore, the steps of obtaining the atmospheric interference phase sample points in step S40 include:

[0034] S41. Subspace-divide the interference phase in the azimuth direction by an angle θ Azm for the image detection range;

[0035] S42. Space-grid-divide the subspace according to the window size of W x *W y where W x and W yrespectively represent the horizontal and vertical lengths of the time window for subspace division. The interference phases of all the permanent scatterers in each grid are weighted and averaged using the average correlation coefficient to obtain the first mean phase. where the weighting value is r;

[0036] S43. Weight and average the interference phases of all the permanent scatterers in the sub-grid again using the amplitude deviation to obtain the second mean phase. where the weighting value is 1 - D A ;

[0037] S44. Take as the approximate value of the atmospheric interference phase at the grid center point, and simultaneously calculate the radial distance R from this grid center point to the radar w , which is the sample point of the atmospheric interference phase. Here, w is the spatial grid number;

[0038] S45. Traverse all the divided subspaces according to the steps S42 to S44.

[0039] Furthermore, the angle θ in the step S41 Azm is set to 30°.

[0040] Furthermore, the steps for obtaining the available sample points in the step S50 include:

[0041] S51. Statistically calculate the percentage P of the number of permanent scatterers in the divided grids of each subspace accounting for the total number of pixels in the current grid w (%).

[0042] S52. Set the third threshold T 3 . When the P w ≥ T 3 , it is the available sample point of the atmospheric interference phase.

[0043] Furthermore, the third threshold T in the step S52 3 is set to 10.

[0044] Furthermore, the step S60 includes:

[0045] S61. Model the atmospheric interference phase into a linear model varying with the slant range:

[0046]

[0047] where β 0 is the constant component, β 1 is the linear coefficient related to the slant range, and λ is the wavelength of the system transmitted signal;

[0048] S62. Establish a system of linear equations:

[0049]

[0050] Among them, β = [β 0 β 1 T , ε = [ε 1 , ε 2 , …, ε w T is a random error vector, and T represents the matrix transpose;

[0051] S63. Estimate β using the least squares method, and we can get:

[0052]

[0053] Then the estimated value of the atmospheric interference phase is:

[0054]

[0055] Among them, The difference between is the compensated interference phase;

[0056] S64. Repeat the steps S61 to S63, and the estimation of the atmospheric interference phase for the entire detection space can be completed.

[0057] Compared with the prior art, the ground-based circular arc interferometric synthetic aperture radar atmospheric interference phase correction method provided by the present invention has at least the following beneficial effects:

[0058] The present invention introduces the concept of permanent scatterers, screens the permanent scatterers through the mean value of the correlation coefficient and the amplitude deviation, extracts the stable permanent scatterers with the deformation phase approximately zero, and uses the correlation coefficient and the amplitude deviation to perform weighted averaging on the interference phase of the stable permanent scatterers as the sample points of the atmospheric interference phase. At the same time, the detection area is divided into subspaces in the azimuth direction, and the subspaces are further divided into spatial grids. The equivalent sample points of the atmospheric interference phase are determined by using the sample points of the atmospheric interference phase within the grid, and then the least squares estimation of the atmospheric interference phase model parameters is completed. Without introducing external auxiliary data, the effective correction of the atmospheric interference phase is carried out relying on the interference phase itself; fully considering the two-dimensional spatial variation characteristics of the atmospheric interference phase, the problem of under-correction of the atmospheric interference phase in the azimuth direction is solved through subspace and grid division; the method flow is simple and easy to implement, and has strong adaptability and robustness, and has a good correction effect on the atmospheric interference phase, meeting the requirements of engineering practice. Description of the Drawings​​

[0059] To more clearly illustrate the solution of the present invention, the following will give a brief introduction to the drawings required in the description of the embodiments. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0060] Figure 1 It is a flowchart of a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar provided by an embodiment of the present invention;

[0061] Figure 2 It is a flowchart of a method for screening permanent scatterers in a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar provided by an embodiment of the present invention;

[0062] Figure 3 It is a flowchart of screening sample points of atmospheric interference phase in a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar provided by an embodiment of the present invention. Detailed implementation manners

[0063] To facilitate the understanding of the present invention, the following will describe the present invention more comprehensively with reference to the relevant drawings. The preferred embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure content of the present invention more thorough and comprehensive.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0065] Such as Figure 1The following is a flowchart of a method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar provided by an embodiment of the present invention. This method takes into account the two-dimensional spatial variation characteristics of the atmospheric interference phase. By introducing the concept of Permanent Scatters (PS), the PS points are screened using the mean of the correlation coefficient and the amplitude deviation. Stable PS points with approximately zero deformation phase are extracted, and the weighted average of the interferometric phases of the stable PS points using the correlation coefficient and the amplitude deviation is used as the sample points of the atmospheric interference phase. At the same time, the detection area is divided into subspaces in the azimuth direction, and each subspace is further divided into spatial grids. The equivalent sample points of the atmospheric interference phase are determined using the sample points of the atmospheric interference phase within the grids, and then the least squares estimation of the model parameters of the atmospheric interference phase is completed. Without introducing external auxiliary data, the effective correction of the atmospheric interference phase is carried out relying on the interferometric phase itself.

[0066] The specific steps are as follows:

[0067] S10. When the ground-based circular arc interferometric synthetic aperture radar system actually operates, affected by the atmospheric interference phase error and the noise phase error, the interferometric phase satisfies the following model:

[0068]

[0069] Where, is the interferometric phase between the main image u and the slave image v, is the deformation phase, is the atmospheric interference phase, is the noise phase;

[0070] S20. Filter the interferometric phase to eliminate the phase noise;

[0071] In view of the high requirements for noise suppression in the deformation measurement of the ground-based circular arc interferometric synthetic aperture radar, the noise phase filtering algorithm needs to have both efficiency and performance. The weighted circular period median filtering algorithm is currently the most widely used ground-based synthetic aperture radar phase filtering algorithm. The following output can be obtained after filtering the interferometric phase in expression (1):

[0072]

[0073] Where, (kn, km) is the coordinate serial number of the PS point in the interferometric image, n is the abscissa of the interferometric phase to be filtered, m is the ordinate of the interferometric phase to be filtered, and the weighting coefficient is Where, M is the length of the filtering window on the horizontal axis, N is the length of the filtering window on the vertical axis, (median is the median filtering),

[0074] S30. Screen out the Permanent Scatters (PS) from the interferometric phase according to the statistical characteristic quantities, where the permanent scatterers are the control points with the interferometric phase approximately zero, that is Therefore, the interferometric phase after filtering in step S20 can be expressed as:

[0075]

[0076] In this step, the screening characteristic quantities of the permanent scatterers mainly include the correlation coefficient and the amplitude deviation, and the screening of stable permanent scatterers is completed through secondary screening;

[0077] S40. Perform subspace partitioning on the interferometric phase obtained in step S20 in the azimuth direction, then divide the subspace into grids according to the set window size, and perform phase weighted average processing on the permanent scatterers in the grid according to the correlation coefficient and the amplitude deviation, and use it as the atmospheric interference phase sample point of the grid center point;

[0078] S50. Screen the atmospheric interference phase sample points in each subspace obtained in step S40 to obtain available sample points;

[0079] S60. Use the available sample points to estimate the atmospheric interference phase error model parameters of the corresponding subspace, and then obtain the atmospheric interference phase error of the corresponding detection subspace;

[0080] S70. According to steps S10 to S60, the deformation phase of the entire detection space after compensation can be obtained.

[0081] As Figure 2 shown is the flowchart of the permanent scatterer screening method in step S30 of a ground-based circular arc interferometric synthetic aperture radar atmospheric interference phase correction method provided by an embodiment of the present invention, which specifically includes the following steps:

[0082] S31: Since the correlation coefficient can reflect the signal-to-noise ratio of the interferometric phase diagram, a reasonable correlation coefficient threshold (set to [0.9, 1] in this embodiment) can be set, and the scatterers with a larger mean correlation coefficient are selected as the candidate point set of the permanent scatterers. The correlation coefficient can be calculated according to the amplitude and phase information of the pixel point to be calculated and its surrounding neighborhood points, as shown in expression (4):

[0083]

[0084] Among them, M(i, j) and S(i, j) are the master and slave complex scattering images that make up the interferometric phase, * represents complex conjugate, W1 is the size of the first sliding window, and w2 is the size of the second sliding window;

[0085] S32. Use N pFor the amplitude time series complex scattering images, N p -1 correlation coefficient matrices can be calculated, and then the average correlation coefficient can be calculated by Expression (5):

[0086]

[0087] where q represents the interference phase of the q-th frame, and γ q represents the correlation coefficient of the q-th frame;

[0088] S33: For high-quality pixel points, the amplitude time series information can characterize their radiation stability, and the amplitude deviation is a direct characterization quantity of the amplitude time series information. Therefore, a reasonable threshold (set to [0, 0.1] in this embodiment) can be set, and the amplitude deviation is used to perform secondary screening of the permanent scatterers;

[0089] S34. Calculate the amplitude deviation D A of the amplitude time series complex scattering image:

[0090]

[0091] where δ A represents the standard deviation of the pixel amplitude time series, and m A represents the mean of the pixel amplitude time series. It can be seen from Expression (6) that the smaller the amplitude deviation of the pixel, the higher its radiation stability, and the smaller the corresponding phase change, that is, the interference phase is closer to zero.

[0092] As Figure 3 shown in the flowchart of step S40 for screening the atmospheric interference phase sample points in the ground-based circular arc interferometric synthetic aperture radar atmospheric interference phase correction method provided by the embodiment of the present invention, it specifically includes the following steps:

[0093] S41: Since the ground-based circular arc interferometric synthetic aperture radar usually operates in a large-range scanning mode (large angle, long distance), the atmospheric interference phase error has two-dimensional spatial variation characteristics. However, within a local small area, the influence of the atmosphere is still consistent. Therefore, the detection space can be considered to be divided into subspaces in the azimuth direction at an angular interval θ Azm (set to 30° in this embodiment), and one-dimensional atmospheric error estimation and compensation in the range direction are performed for each subspace separately;

[0094] S42: The subspace is divided into spatial grids according to a window size of W x *W y (set to 30*30 in this embodiment), where W x and W yrespectively represent the horizontal and vertical lengths of the time window for subspace division. The interferometric phases of all permanent scatterers in each grid are weighted and averaged using the average correlation coefficient to obtain the mean phase where the weighting value is r;

[0095] S43: The interferometric phases of all permanent scatterers in the sub-grid are weighted and averaged again using the amplitude deviation to obtain the mean phase where the weighting value is 1 - D A ;

[0096] S44: Take as the approximate value of the atmospheric interference phase at the grid center point, and simultaneously calculate the radial distance R from the grid center point to the radar w , which is the sample point of the atmospheric interference phase. Here, W is the spatial grid number;

[0097] S45: Traverse all the divided subspaces according to steps S42 to S44 to obtain the set of atmospheric interference phase points corresponding to each subspace.

[0098] To improve the accuracy of the atmospheric interference phase parameters, it is necessary to screen the quality of the equivalent atmospheric sample points. Step S50 in the ground - based circular - arc interferometric synthetic aperture radar atmospheric interference phase correction method provided by the embodiment of the present invention gives the screening process of the available equivalent atmospheric interference phase sample points, which specifically includes the following steps:

[0099] S51: Statistically calculate the percentage P of the number of permanent scatterers in the divided grids of each subspace accounting for the total number of pixels in the current grid w (%), and the arithmetic formula is shown in expression (7):

[0100] P w = N ps / (W x * W y ) * 100 (7)

[0101] S52: Set the threshold T 3 (set to 10 in this embodiment). When P w ≥ T 3 , it is the available equivalent atmospheric interference phase sample point; otherwise, directly discard it.

[0102] Traverse all the spatial grids of the current subspace to obtain the set of available equivalent atmospheric interference phase sample points.

[0103] Since subspace division is performed in the azimuth direction, in each subspace, the atmosphere has no spatial variability in the azimuth direction, and the atmospheric interference phase error is only related to the slant range. Then, by taking the steps in step S60 of this embodiment, the estimation and compensation of the atmospheric interference phase error can be completed. The specific steps are as follows:

[0104] S61: Construct a linear model of the atmospheric interference phase varying with the slant range, as shown in expression (8):

[0105]

[0106] where, β 0 is the constant component, and β 1 is the linear coefficient related to the slant range;

[0107] S62. Establish a system of linear equations:

[0108]

[0109] where, β = [β 0 β 1 T , ε = [ε 1 , ε 2 , …, εw] T is the random error vector, and T represents the matrix transpose;

[0110] S63. Use the least squares method to estimate β, and we can get:

[0111]

[0112] Then the estimated value of the atmospheric interference phase is:

[0113]

[0114] where, The difference between and

[0115] ​A method for correcting the atmospheric interference phase of a ground-based circular arc interferometric synthetic aperture radar described in the above embodiments introduces the concept of permanent scatterers, screens the permanent scatterers through the mean of the correlation coefficient and the amplitude deviation, extracts stable permanent scatterers with approximately zero deformation phase, and uses the correlation coefficient and the amplitude deviation to perform weighted averaging on the interferometric phase of the stable permanent scatterers as the sample points of the atmospheric interference phase. At the same time, the detection area is divided into subspaces in the azimuth direction, and each subspace is further divided into spatial grids. The equivalent sample points of the atmospheric interference phase are determined by using the sample points of the atmospheric interference phase within the grids, and then the least-squares estimation of the model parameters of the atmospheric interference phase is completed. Without introducing external auxiliary data, the effective correction of the atmospheric interference phase is carried out relying on the interferometric phase itself; the two-dimensional spatial variation characteristics of the atmospheric interference phase are fully considered, and the problem of under-correction of the atmospheric interference phase in the azimuth direction is solved through subspace and grid division; the method flow is simple and easy to implement, and has strong adaptability and robustness, and has a good correction effect on the atmospheric interference phase, meeting the requirements of engineering practice.

[0116] Obviously, the embodiments described above are only the preferred embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the drawings, but do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing specific embodiments, or perform equivalent replacements on some of the technical features. Any equivalent structure using the content of the specification and drawings of the present invention, directly or indirectly applied to other related technical fields, is similarly within the scope of the patent protection of the present invention.

Claims

1. A method for atmospheric interference phase correction of ground-based circular arc interferometric synthetic aperture radar. It is characterized in that The following steps are involved: S10, constructing a ground-based circular arc interferometric synthetic aperture radar differential interferometry phase model, wherein the phase model is specifically: Among them, is the interference phase between the main image u and the slave image v, is the deformation phase, is the atmospheric interference phase, is the noise phase; S20, filtering the interference phase to eliminate phase noise; S30, selecting a permanent scattering point from the interference phase according to a statistical characteristic quantity, wherein the permanent scattering point is a control point where the interference phase is approximately zero; S40, performing subspace division on the interference phase obtained in step S20 in the azimuth direction, and then dividing the subspace into grids according to a set window size, performing phase weighted averaging processing on the permanent scattering points in the grid according to the correlation coefficient and the amplitude deviation, and taking the atmospheric interference phase sample points as the grid center points; The step of obtaining the atmospheric interference phase sample points in step S40 includes: S41. Subdivide the image detection range into subspaces according to the angle θ in the azimuth direction of the interference phase Azm for the sub - space division of the image detection range; S42. Divide the subspace into a spatial grid according to the window size of W x *W y where W x and W y represent the horizontal and vertical lengths of the window during subspace division respectively. Use the average correlation coefficient to perform weighted averaging on the interferometric phases of all the permanent scatterers in each grid to obtain the first mean phase where the weighting value is S43. Using the amplitude deviation to perform weighted averaging on the interference phases of all the permanent scatterers within the sub-grid again to obtain a second mean phase where the weighting value is 1-D A ; S44. Take the approximate value of the atmospheric interference phase as the center point of the grid, and simultaneously calculate the radial distance R from this grid center point to the radar w , as the sample point of the atmospheric interference phase, where W is the spatial grid number; S45, traversing all divided subspaces according to steps S42 to S44; S50, screening the atmospheric interference phase sample points in each subspace obtained in step S40 to obtain available sample points; S60, estimating the atmospheric interference phase error model parameters of the corresponding subspace using the available sample points, and then obtaining the atmospheric interference phase error of the corresponding detection subspace; S70: Obtaining the deformation phase of the entire detection space after compensation.

2. According to claim 1, a method for correcting atmospheric interference phase of ground-based circular arc interferometric synthetic aperture radar, It is characterized in that The filtering process in step S20 is weighted circular periodic median filtering, and the weighting coefficient is: where (kn, km) are the coordinate indices of the permanent scatterer in the interference image, n is the abscissa of the interference phase to be filtered, and m is the ordinate of the interference phase to be filtered. where M is the length of the filtering window on the horizontal axis and N is the length of the filtering window on the vertical axis.

3. The atmospheric interference phase correction method for ground-based circular arc interferometric synthetic aperture radar according to claim 1, It is characterized in that The step of screening out permanent scattering points in step S30 includes: S31, calculate the correlation coefficient γ according to the master-slave complex scattering image, the formula is as follows: Wherein, M(i, j) and S(i, j) are respectively the main complex scattering image and the slave complex scattering image constituting the interference phase, * represents complex conjugation, w1 is the first sliding window size, and w2 is the second sliding window size; S32. Use N p temporal complex scattering images of the images to calculate the average correlation coefficient. The formula is as follows: where q represents the interference phase of the q-th frame, and γ q represents the correlation coefficient of the q-th frame; S33. Set the first threshold T 1 , and screen out the pixel points of the as candidate points for the permanent scatter points; S34. Calculate the amplitude deviation D of the time-series complex scattering image A , and the formula is as follows: where δ A represents the standard deviation of the pixel amplitude time series, and m A is the mean of the pixel amplitude time series; S35. Set the second threshold T 2 , and screen out the permanent scatterers with D A ≤T 2 from the candidate point set of the permanent scatterers obtained in the step S31.

4. According to claim 3, a method for correcting atmospheric interference phase of ground-based circular arc interferometric synthetic aperture radar, It is characterized in that The first threshold T in step S32 1 is set to 0.9, and the second threshold T 2 is set to 0.

1.

5. The atmospheric interference phase correction method for ground-based circular arc interferometric synthetic aperture radar according to claim 1, It is characterized in that The angle θ in the step S41 Azm is set to 30°.

6. The atmospheric interference phase correction method for ground-based circular arc interferometric synthetic aperture radar according to claim 1, It is characterized in that The step of obtaining the available sample points in step S50 includes: S51. Statistically calculate the percentage P of the number of the permanent scatterers in the divided grids of each subspace to the total number of pixels in the current grid w ; S52. Set the third threshold T 3 , when the P w ≥T 3 , it is the available atmospheric interference phase sample point.

7. The atmospheric interference phase correction method for ground-based circular arc interferometric synthetic aperture radar according to claim 6, It is characterized in that The third threshold T described in step S52 3 is set to 10.

8. The atmospheric interference phase correction method for ground-based circular arc interferometric synthetic aperture radar according to claim 1, It is characterized in that The step S60 comprises: S61, modeling the atmospheric interference phase into a linear model that varies with slant range: where β 0 is the constant component, and β 1 is the linear coefficient related to the slant range, and λ is the wavelength of the system transmitted signal; S62. Establish a linear equation system: Among them, β = [β 0 β 1 T , ε = [ε 1 , ε 2 , …, ε w T is a random error vector, and T represents matrix transpose;​​ S63. Using the least squares method to estimate β, we can get: Then the atmospheric interference phase estimation value is as follows: Among them, and the difference between them is the compensated interference phase; S64. Repeat the steps S61 to S63, and the estimation of the atmospheric interference phase for the entire detection space can be completed.

Images