A method for splicing time series results based on homologous SAR data

Abstract

The embodiment of the present invention discloses a method for splicing time-series InSAR results based on homologous SAR data, which obtains deformation monitoring data of each sub-block through InSAR technology; obtains complete deformation information of the study area using an image splicing method; and finally obtains a complete ground deformation field. The present invention innovatively proposes to use the terminal data as a reference to identify and correct system errors when splicing data, avoid the transmission of system errors when splicing long strips of data, improve the traditional method of obtaining points at the same position in overlapping areas, and make the calculation of benchmark deviations more accurate, which is suitable for obtaining continuous, complete, and accurate deformation information in large-scale study areas.

Description

Technical Field

[0001] The present invention relates to the field of time-series InSAR deformation monitoring, and is particularly applicable to the field of time-series InSAR result splicing of homologous SAR data. Background Art

[0002] Land subsidence refers to a geological phenomenon in which the loose soil layer on the surface of the earth's crust is compressed due to natural and human factors, resulting in a decrease in ground elevation. During the period of rapid urbanization in my country, land subsidence has been monitored in many cities. Land subsidence, especially uneven subsidence, has become a geological disaster problem that affects urban planning and economic development.

[0003] Conventional monitoring methods for ground subsidence are mainly based on leveling, layered marking and GPS measurement. This method is not only labor-intensive, costly and time-consuming, but also easily affected by human factors, resulting in reduced monitoring accuracy. Differential InSAR (D-InSAR) technology is widely used in urban surface subsidence monitoring research due to its high spatial resolution, wide monitoring range and all-day and all-weather characteristics. Advanced time-series differential interferometry technology represented by Persistent Scatterer InSAR (PS-InSAR) and Small Baseline Subsets (SBAS-InSAR) can overcome the influence of traditional D-InSAR technology such as temporal and spatial incoherence and atmospheric effects, and effectively improve the monitoring accuracy of surface deformation.

[0004] Time-series InSAR technology uses radar images as data sources to monitor ground subsidence, but the coverage of single-view SAR data is limited, and with the expansion of range and azimuth, the factors that interfere with the accuracy of deformation signal calculation during image processing will also increase. In order to obtain a continuous and accurate deformation field, the present invention first divides the data into sub-blocks with overlapping areas, performs time-series InSAR processing on the sub-blocks, and then uses image stitching technology to obtain a complete surface deformation field. At present, SAR data stitching methods are mostly aimed at simple two-view or three-view data, and the research focus is also placed on the stitching of different track data, ignoring the stitching problem of long strip data on the same track, and is not suitable for the acquisition of complete deformation fields in research areas with large north-south spans. The acquisition of data of points at the same position in the overlapping area is mostly converted into vectors after simple interpolation. Although this method increases the number of points at the same position, too much data increases the redundancy of calculation, and in areas with low coherence, the settlement rate generated only by numerical fitting of surrounding PS points is not accurate, which is easy to cause stitching errors.

[0005] Therefore, in view of the above problems, the embodiment provided by the present invention improves the conventional data splicing method, pre-selects the terminal data as the reference data to ensure the accuracy of the results, and is used to identify and correct the system errors during data splicing to avoid the transmission of system errors. The data acquisition of the same position point only interpolates the data of one scene with more PS points, and uses the latitude and longitude coordinates of the PS points of another scene to extract the same position points. The present invention is particularly suitable for the continuous, complete and accurate acquisition of ground subsidence information in a study area with a large north-south span. Summary of the invention

[0006] An embodiment of the present invention provides a method for splicing time-series InSAR results based on homologous SAR data, the method comprising:

[0007] The SAR data covering the study area is divided into sub-block data with overlapping areas using the block method;

[0008] The PS-InSAR technology is used to process the data of each sub-block, obtain the InSAR results of each sub-block and obtain the deformation information from them;

[0009] Select the SAR data of the terminal sub-block of the same track, use SBAS-InSAR technology to process it to obtain another set of deformation information, cross-validate the accuracy of the deformation information of the terminal sub-block, and use it as reference data;

[0010] Based on the InSAR results of the overlapping areas between sub-blocks of the same track strip, the reference deviation of the sedimentation rate between sub-blocks of the same track is determined according to the information of the same position points, and the reference deviation is averaged and compensated to the data of each adjacent sub-block. The InSAR results of adjacent sub-blocks are spliced, and the system deviation is corrected based on the reference data to complete the splicing of the InSAR results of the same track strip.

[0011] The InSAR results of adjacent tracks are unified into the geographic coordinate system, and the annual sedimentation rate in the adjacent track InSAR detection results is converted into the vertical annual sedimentation rate to eliminate the incident angle error. Based on the information of the same position points of adjacent tracks and the annual sedimentation rate of the same position points, the benchmark deviation of the sedimentation rate between each sub-block of the adjacent tracks is determined, and the benchmark deviation is averaged and compensated to each sub-block of the adjacent tracks to complete the splicing of the InSAR detection results within the study area.

[0012] The leveling point data is used to verify the splicing results and obtain a complete ground deformation field.

[0013] Compared with the prior art, the embodiment of the present invention provides a method for splicing time-series InSAR results based on homologous SAR data, including splicing of same-track data with a larger north-south span and splicing of adjacent-track data with a larger east-west span. It innovatively proposes to use the terminal data as a reference to identify and correct the system error when splicing data, avoid the transmission of system error when splicing long strips of data, improve the traditional method of obtaining the same-position points in the overlapping area, make the calculation of the benchmark deviation more accurate, and is suitable for obtaining continuous, complete, and accurate deformation information in a large-scale study area. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to make the technical solutions and advantages in the embodiments of the present application more clearly understood, the exemplary embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. Obviously, the drawings described below are only multiple embodiments disclosed in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a technical flow chart of the method of splicing time-series InSAR results based on homologous SAR data;

[0016] Figure 2 This is a schematic diagram of the effect of splicing the time-series InSAR results based on the same source SAR data;

[0017] Figure 3 This is a comparison chart of the effects before and after splicing of time-series InSAR results based on homologous SAR data;

[0018] Figure 4 It is a schematic diagram of the splicing method of time-series InSAR results based on homologous SAR data;

[0019] Figure 5 It is a schematic diagram of the method of stitching InSAR results of the same track;

[0020] Figure 6 It is a schematic diagram of the method for stitching InSAR results from adjacent tracks. DETAILED DESCRIPTION

[0021] The scheme provided by the embodiments of this specification is described below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the invention. The described embodiments are only part of the embodiments of this specification, rather than all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0022] The embodiment of the present invention discloses a method for splicing time-series InSAR results based on homologous SAR data. The technical flow chart of the method is as follows: Figure 1 As shown, the specific process is as follows:

[0023] First, determine the scope of the study area, download the SAR data covering the study area, and use the block method to divide the SAR data into sub-blocks with overlapping areas. The specific implementation is as follows:

[0024] 1) Determine the scope of the study area and download SAR data.

[0025] The present invention takes the 10km area on the left and right banks of the Grand Canal as an example to introduce the specific implementation method. The Grand Canal is more than 1,700 kilometers long, runs through 2 cities and 4 provinces in my country, and connects five major water systems. The Sentinel-1A images covering the study area from May 2017 to December 2020 are selected as the data source. A total of 352 SAR data from two tracks are required. The Grand Canal is covered every 8 scenes at the same time, of which track 142 requires 5 scenes and track 69 requires 3 scenes. One scene is selected every month, and the data acquisition time of the two tracks is as consistent as possible to ensure the subsequent processing and splicing process.

[0026] 2) Block-based processing of Sentinel-1 data

[0027] The present invention comprehensively considers the effects of data processing efficiency, orbit error, atmospheric additional phase, etc., firstly divides the SAR data into sub-blocks with overlapping areas by using a block method, and then performs time-series InSAR processing.

[0028] Then, the PS-InSAR technology is used to process the data of each sub-block to obtain the InSAR results of each sub-block and obtain deformation information therefrom. The specific implementation example is as follows:

[0029] Time-series InSAR technology includes a variety of methods, and users can choose according to their needs. In this case, PS-InSAR technology is selected for deformation monitoring. First, considering the influence of temporal and spatial baseline distance and Doppler centroid frequency difference, one image is selected as the main image. After reading POD precise orbit determination ephemeris data and meteorological information, the remaining auxiliary images are respectively registered and resampled with the main image. The registered data set is preprocessed, the average amplitude image and the time-series amplitude deviation index are calculated, the DEM data is read, and the ground control points are selected to calculate the conversion constants for removing the terrain and the horizon phase. The interference processing parameters are set to remove the terrain and the horizon phase, and the multi-view parameters are set to calculate the interference map and the coherence coefficient map. Then, the atmospheric phase analysis is performed, and the effective PS candidate points are selected to construct the analysis pixel network. The strong reflection points with strong coherence and stable reflection characteristics are selected as reference points. After solving the various parameters and the atmospheric delay phase components based on the reference points, the correctness of the function model solution is verified by comparing with the interference phase of each pixel point. Finally, the atmospheric delay phase analysis results obtained previously are read, and the linear deformation, nonlinear deformation, elevation difference, seasonal thermal deformation, etc. are refined and calculated. The threshold is set to select PS points with higher coherence to obtain high-precision PS point deformation analysis results.

[0030] Secondly, the SAR data of the terminal sub-block of the same track is selected, and the SBAS-InSAR technology is used to process to obtain another set of deformation information, cross-validate the deformation information accuracy of the terminal sub-block, and use it as reference data. The specific implementation example is as follows:

[0031] In order to avoid the transmission of system errors when splicing long strips of data, the present invention selects Track142-Frame106 and Track69-Frame94 data as references. After the time-series InSAR processing, the accuracy is verified to ensure the accuracy of the results. A variety of methods can be selected for verification, such as level point verification, cross-validation of the results of multiple InSAR monitoring methods, etc. Level point data is difficult to obtain, so this case selects the cross-validation of SBAS and PS results.

[0032] Thirdly, based on the information of the same position points in the InSAR results of the overlapping areas between the sub-blocks of the same track strip, the reference deviation of the sedimentation rate between the sub-blocks of the same track is determined, the reference deviation is averaged and compensated to the data of each adjacent sub-block, the InSAR results of the adjacent sub-blocks are spliced, and the system deviation is corrected based on the reference data to complete the splicing of the InSAR results of the same track strip. The specific implementation is as follows:

[0033] The present invention uses the annual average sedimentation rate value of the PS point in the overlapping area for splicing, first splicing the long strip data of the same track, and then splicing the adjacent track data. The main reasons for the difference in sedimentation rate results monitored by different images of the same position point in the overlapping area include two aspects: the deviation of the PS point position and the deviation of the PS sedimentation of the same object. Due to the differences in coherence and phase in the independent processing of different strip / track data, the high coherence points (PS) in the overlapping area cannot be guaranteed to be identified at the same time. In addition, since SAR data is acquired based on slant range, each imaging data point in the image has a different incident angle, which causes the PS point to represent different positions of the same object. The deviation of the PS point sedimentation is mainly caused by the difference in reference points, the incident angle, the atmospheric phase difference, and other errors such as DEM error and noise error.

[0034] For the deviation of the PS point position, the present invention corrects it by an interpolation method. The commonly used interpolation method is to convert two sets of data into vector data after interpolation respectively, and extract the same position points, but this method is not only easy to cause data redundancy, but also in the area with low coherence, the sedimentation rate generated by the numerical fitting of the surrounding PS points alone is not accurate, and it is easy to cause splicing errors. Therefore, the present invention selects a group of data with a large number of PS points, and interpolates with the PS point as the center. Considering the influence of the distance between the PS point and the reference point, the inverse distance weighted interpolation method (IDW) method is selected during interpolation. The selection of PS points is based on pixels, and radar pixels and ground pixels show a corresponding relationship. Therefore, the interpolation window size should be consistent with the ground pixel size, so as to generate a continuous surface deformation field. Then, with a data set with fewer PS points as a reference, the deformation rate result of the corresponding position of the surface deformation field is extracted by longitude and latitude coordinates, and used as the data of the same position points in the overlapping area.

[0035] As for the deviation of sedimentation, PS-InSAR technology itself can eliminate some atmospheric effects and DEM errors. Therefore, when stitching data, the deviation of sedimentation caused by reference points, incident angle and other factors is mainly considered.

[0036] The deformation rate difference △d of the PS point of the same feature in the overlapping area i It can be expressed as:

[0037] △d i =d i (n)-d i (n-1) = △D(ref) + △d i (θ)+△d i_error (3-1)

[0038] In formula (3-1), ΔD(ref) is the difference in deformation rate caused by different reference points, △d i(θ) is the difference in deformation rate caused by different incident angles. Therefore, the SAR data splicing problem is transformed into solving △D(ref) and △d by using the deformation rate difference i (θ) problem. ΔD(ref) This paper solves it by calculating the annual average sedimentation rate of the same position points in the overlapping area, using statistical analysis methods to collect information on the same position points, and using formula (3-2) to calculate the overall deviation of the overlapping area.

[0039]

[0040]

[0041] In formula (3-2), Δv off represents the reference deviation between sub-blocks, N is the number of points at the same position in the overlapping area, and v ref,i and v s,i is the linear deformation rate of high coherence point target i between different sub-blocks, v s,i is the linear deformation rate of the high coherence point i after benchmark compensation.

[0042] During data processing, each sub-block has an independent reference benchmark, and the activity of each reference point is usually unknown. Since they have the same track number, the error caused by the difference in incident angle can be ignored, so the splicing of the same track long strip data requires compensation for the benchmark deviation between different sub-block data. Since the Beijing-Hangzhou Grand Canal has a large north-south span, it is easy to cause the transmission of systematic errors during splicing. Therefore, two sub-blocks are selected in the middle and south of the study area as reference data. The selection of reference points is strictly controlled during processing, and the accuracy of the monitoring results is verified so that they can be used as a benchmark to correct the systematic deviation during subsequent splicing.

[0043] The other data results are interpolated to obtain the same-position points in the overlapping area, and the information of the same-position points is substituted into the formula to obtain the benchmark deviation, and then spliced ​​from north to south. The basic idea is to extract the information of the same-position points in the overlapping area of ​​the sub-blocks through interpolation, and based on the average annual sedimentation rate, substitute it into formula (3-2) to obtain the benchmark deviation Δv between the sub-blocks. off. Due to the lack of prior experience, it is impossible to specify a sub-block as the reference image during stitching. Therefore, the obtained benchmark deviation is evenly distributed to the two sub-blocks for compensation, and the stitching is performed from north to south. After stitching, it is compared with the reference data results to set the threshold. If the deviation is within the threshold range, it is judged to be valid and the stitching is completed. If it exceeds the threshold, it is necessary to find the cause of the error again, reprocess, revise, and stitch again until it is within the allowable error range. The setting of the threshold should refer to the resolution of the image and the needs of the user. When the resolution is low, the threshold can be set larger. The SAR data used in this case is only of medium resolution, so the average difference threshold is set to ±5mm / a, and the mean error threshold is set to ±3mm / a. The average difference between the stitched data and the reference data in this case is 0.07mm / a and 3.83mm / a, respectively, and the mean error is 1.19mm / a and 1.37mm / a, respectively, all within the threshold range, and the stitching result is judged to be accurate.

[0044] Next, based on the InSAR results of the overlapping areas between the sub-blocks of the same track strip, according to the information of the same position points, the reference deviation of the sedimentation rate between the sub-blocks of the same track is determined, the reference deviation is averaged and compensated to the data of each adjacent sub-block, the InSAR results of the adjacent sub-blocks are spliced, and the system deviation is corrected based on the reference data to complete the splicing of the InSAR results of the same track strip. The specific implementation is as follows:

[0045] For the difference in deformation rate △d caused by different incident angles i (θ), the data obtained by PS-InSAR is the component in the radar line of sight. The one-dimensional displacement of the same feature projected to different line of sight will be different, but the displacement of multi-platform data projected to the vertical direction is the same. Therefore, the present invention ignores the difference in the horizontal direction and uses formula (3-4) to convert the line of sight deformation to the vertical direction to eliminate the error caused by the difference in incident angle.

[0046]

[0047] In formula (3-4), D VA(D) is the vertical deformation rate; D los is the line-of-sight deformation rate, and θ is the incident angle.

[0048] The SAR data acquired under different orbits have different incident angles and are located in independent radar coordinate systems. PS-InSAR processing is based on the radar coordinate system. Due to different geometric imaging, the sedimentation rate results of PS points on different orbits are also located in different radar coordinate systems. Therefore, the coordinate system needs to be unified before stitching. Common coordinate systems include radar coordinate system and geographic coordinate system. Users can choose according to their needs. In this case, the geographic coordinate system is selected.

[0049] The basic idea of ​​adjacent track splicing is to use formula (3-4) to convert the annual average settlement rate in the line of sight into the annual average settlement rate in the vertical direction to eliminate the incident angle error. The interpolation method is used to obtain the same position points in the overlapping area. The information of the same position points is used to calculate the benchmark deviation using formulas (3-2) and (3-3). The benchmark deviation is evenly distributed to the two track data for compensation. Finally, the accuracy of the spliced ​​data is verified and the splicing is completed.

[0050] Finally, the level point data is used to verify the splicing results and obtain a complete ground deformation field. The specific implementation example is as follows:

[0051] Figure 2 This is the complete ground subsidence information after splicing the Beijing-Hangzhou Grand Canal. Overall, the deformation field is continuous without obvious splicing marks. Figure 3 This is a schematic diagram of the effect of stitching the time-series InSAR results based on the same-source SAR data. It can be seen from the figure that the average annual sedimentation rate of Track142 in the overlapping area is significantly higher than that of Track69 before stitching correction. After processing using the stitching technology, except for differences in a small number of areas, the sedimentation rates in other areas are basically the same.

[0052] Table 1 shows the verification results of the leveling point data. The average error between the leveling point and the InSAR monitoring data before splicing is -5.03 mm / a, and the average error after splicing is -2.24 mm / a. The difference between the leveling data and the InSAR monitoring data after splicing is within 10 mm / a. The maximum error is -7.71 mm / a at P9, the minimum error is -1.04 mm / a at P1, and the median error is 3.25 mm / a.

[0053]

[0054] The embodiment of the present invention discloses a method for splicing time-series InSAR results based on homologous SAR data. The method is shown as follows: Figure 4 As shown, the specific method is as follows:

[0055] S410: Divide the SAR data covering the study area into sub-block data with overlapping areas by using a block method.

[0056] S420: Obtaining deformation monitoring results of each sub-block

[0057] The PS-InSAR technology is used to process the SAR data of each sub-block and calculate the coherence coefficient map and interference map of the sub-block data. Candidate points with high coherence and strong reflective characteristics are selected as reference points in the analysis pixel point network established by valid PS candidate points. Atmospheric phase analysis is performed, thresholds are set, and permanent scatterers with high coherence are selected to determine the continuous and complete deformation information of each sub-block.

[0058] S430: Verify the precision and accuracy of the reference data deformation monitoring results

[0059] The SAR data of the terminal sub-block of the same track were selected and processed using SBAS-InSAR technology to obtain another set of deformation monitoring information, cross-validate the accuracy of the PS-InSAR deformation results, and use it as reference data.

[0060] S440: Complete the splicing of InSAR results of long strips on the same track

[0061] The same track InSAR result splicing method is as follows Figure 5 As shown, the specific implementation is as follows:

[0062] S501: Taking track 1 as an example, first select sub-block 1 and sub-block 2 at the head end in track 1, select a group of data with a large number of permanent scatterers in the overlapping area of ​​the two sub-blocks as a benchmark, use the inverse distance weighted interpolation method to generate a continuous surface deformation field, and then use a group of data with a small number of permanent scatterers as a reference, extract the deformation rate results of the corresponding positions of the surface deformation field through longitude and latitude coordinates, and use them as the data of the same position points in the overlapping area.

[0063] Then, based on the annual sedimentation rate of the same position point of sub-block 1 and sub-block 2, the reference deviation V1 between sub-block 1 and sub-block 2 is determined, and then V1 is averaged and compensated to the two sub-block data to complete the splicing of the InSAR results of the two sub-blocks;

[0064] S502: The spliced ​​InSAR results are used as new sub-block data and spliced ​​with sub-block 3, and the reference deviation V2 of the sedimentation rate between the two sub-blocks is also determined, and then V2 is averagely compensated to the two sub-block data until they are spliced ​​to the terminal sub-block 5;

[0065] S503: Taking the data of sub-block 5 as the benchmark, set the average difference threshold to ±5mm / a and the mean error threshold to ±3mm / a to judge the InSAR results of the sub-blocks on the same track after splicing; if the deviation between the data of the spliced ​​sub-block and the data of sub-block 5 is within the threshold range, it is judged to be valid and the splicing is completed; if it exceeds the threshold, it is necessary to find the cause of the error again, reprocess, revise and splice again until it is within the allowable error range, correct the system deviation and complete the data splicing of the long strip of track 1.

[0066] The stitching method of the long strip InSAR results of track 2 is the same as that of track 1.

[0067] S450: Completed the stitching of InSAR results of adjacent track strips

[0068] The same track long strip InSAR image stitching method is as follows Figure 6 As shown, the specific implementation is as follows:

[0069] S601: Unify the spliced ​​long strip InSAR results of track 1 and track 2 into the geographic coordinate system, convert the annual subsidence rate in the line of sight into the annual subsidence rate in the vertical direction, and eliminate the incident angle error;

[0070] S602: Similar to the method for stitching long strip InSAR results of the same track, first determine the reference deviation V5 between the overlapping sub-blocks of track 1 and track 2.

[0071] S603: V5 is averagely compensated to different sub-block data to complete data splicing of adjacent track long strips.

[0072] S460: Verify the InSAR stitching results of track 1 and track 2 to obtain a complete ground deformation field

[0073] The precision and accuracy of the splicing results are verified using leveling point data to obtain a complete ground deformation field for a long strip.

[0074] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for splicing time-series InSAR results based on homologous SAR data, comprising: The SAR data covering the study area is divided into sub-block data with overlapping areas using the block method; The PS-InSAR technology is used to process the data of each sub-block, obtain the InSAR results of each sub-block and obtain deformation information from them; Select the SAR data of the terminal sub-block of the same track, use SBAS-InSAR technology to process it to obtain another set of deformation information, cross-validate the accuracy of the deformation information of the terminal sub-block, and use it as reference data; Based on the InSAR results of the overlapping areas between sub-blocks of the same track strip, according to the information of the same position points, a group of data with a large number of permanent scatterers is selected as the benchmark, and the continuous surface deformation field is generated using the inverse distance weighted interpolation method. Then, a group of data with a small number of permanent scatterers is used as a reference, and the deformation rate results of the corresponding positions of the surface deformation field are extracted through the longitude and latitude coordinates, which are used as the data of the same position points in the overlapping area. Based on the annual settlement rate of the same location point, the benchmark deviation between sub-blocks is determined; The benchmark deviation is averaged and compensated to the data of each adjacent sub-block, and the InSAR results of adjacent sub-blocks are spliced. The system deviation is corrected based on the reference data, and the InSAR results of the same track strip are spliced. The InSAR results of adjacent tracks are unified into the geographic coordinate system, and the annual sedimentation rate in the adjacent track InSAR detection results is converted into the vertical annual sedimentation rate to eliminate the incident angle error. Based on the information of the same position points of adjacent tracks and the annual sedimentation rate of the same position points, the benchmark deviation of the sedimentation rate between each sub-block of the adjacent tracks is determined, and the benchmark deviation is averaged and compensated to each sub-block of the adjacent tracks to complete the splicing of the InSAR detection results within the study area. Use leveling point data to verify the splicing results and obtain a complete ground deformation field; The mathematical expression of the reference deviation is: Where Δv off represents the reference deviation between sub-blocks, N is the number of points in the same position in the overlapping area, and v ref,i and v s,i is the linear deformation rate of high coherence point target i between different sub-blocks, v s,i is the linear deformation rate of the high coherence point i after benchmark compensation.

2. The method according to claim 1, characterized in that The selection of permanent scatterers is based on pixels, and the interpolation window size is consistent with the ground pixel size.

3. The method according to claim 1, characterized in that The InSAR detection results of stitching adjacent sub-blocks include: Select two sub-blocks at the head end of the same track and determine the reference deviation of the settlement rate between the sub-blocks on the same track; The benchmark deviation is averagely compensated to the two sub-block data to complete the splicing of the InSAR detection results of the two sub-blocks; The spliced ​​InSAR detection results are used as new sub-block data and spliced ​​with the next sub-block until they are spliced ​​to the terminal sub-block.

4. The method according to claim 1, characterized in that The correction of system deviation based on reference data includes: The average difference threshold and the mean error threshold are set. If the deviation between the InSAR detection results of the spliced ​​sub-blocks on the same track and the reference data is within the range of the average difference threshold and the mean error threshold, the spliced ​​data is judged to be valid and the splicing is completed. If it exceeds the threshold range, it is necessary to find the cause of the error again, reprocess, revise, and splice until it is within the allowable error range.

5. The method according to claim 1, characterized in that The mathematical expression for eliminating the incident angle error is: Among them, D VA(D) is the vertical deformation rate; D los is the line-of-sight deformation rate, and θ is the incident angle.

Images