Multi-platform and multi-track InSAR observed quantity splicing method based on unconstrained adjustment

Abstract

The invention discloses a multi-platform and multi-track InSAR observed quantity splicing method based on unconstrained adjustment, and the method comprises the steps: carrying out the polynomial modeling of the difference between homonymy points in an overlapping region, taking the InSAR coherence as a weight, and solving the parameters of a polynomial model through a least square method; a polynomial model is used for correcting the InSAR observed quantity, and then fusion among multiple observations is carried out, so that high-precision integral splicing of the multi-platform and multi-track InSAR observed quantity is realized. According to the method, reliable data support with a unified space-time reference is provided for large-range surface deformation monitoring by improving the splicing precision of the multi-platform and multi-track InSAR observed quantity.

Description

Technical Field

[0001] The invention relates to a multi-platform and multi-track InSAR observational quantity splicing method based on unconstrained adjustment, and belongs to the field of radar remote sensing. Background Art

[0002] Synthetic Aperture Radar (InSAR) is an Earth observation technique based on radar image phase information. It extracts topographic and surface deformation information by interferometrically analyzing SAR (Synthetic Aperture Radar) images taken over the same area at different time periods. Compared to traditional ground-based measurement methods that rely on point observations, spaceborne InSAR technology offers advantages such as wide coverage and high monitoring accuracy. It provides highly accurate and quantifiable information for the detection and monitoring of geological hazards such as ground subsidence.

[0003] The increasing number of SAR satellites, both domestically and internationally, has provided a rich data source for InSAR (InSAR) surface deformation monitoring. However, due to the limited spatial coverage of SAR satellite data from a single platform and orbit, the need for comprehensive InSAR surface deformation information over a wide area requires the combined use of multiple platforms and orbits. Consequently, the need for splicing InSAR observations from multiple platforms and orbits is increasing.

[0004] InSAR observations include differential interferograms, cumulative deformation, average deformation rate, and interferometric coherence results. During the splicing of multi-platform, multi-orbit InSAR observations, factors such as residual orbit error, atmospheric delay error, and differences in radar signal incidence angle and unwrapping reference points can lead to discrepancies between InSAR observations within overlapping regions across different platforms and orbits. This contradicts the theoretical assumption that observations from points of similar origin should be equal. Therefore, adjustment should be performed on the InSAR observations during the splicing process. Currently, commonly used multi-platform, multi-orbit splicing methods include the global additive constant splicing method and the sequential splicing method for multiple observations. However, these methods fail to achieve unified temporal and spatial datums and global splicing of multi-platform, multi-orbit InSAR observations, resulting in significant limitations in observation accuracy.

[0005] In summary, how to minimize the differences and reduce the stitching errors is a key issue in the application of InSAR technology. Summary of the Invention

[0006] To solve the above problems, the present invention proposes a new multi-platform and multi-track InSAR observation splicing method based on unconstrained adjustment.

[0007] To achieve the above object, the solution of the present invention is a multi-platform, multi-track InSAR observation splicing method based on unconstrained adjustment, which comprises the following steps:

[0008] Step 1: Obtain multi-platform, multi-orbit SAR satellite data for a set observation period in the study area, perform interferometric processing and time series analysis, and obtain multiple InSAR observations;

[0009] Step 2: Preprocess the multiple InSAR observations obtained in step 1;

[0010] Step 3: Based on the preprocessing results of step 2, determine the overlapping area of InSAR observations and select the same-name points using InSAR coherence;

[0011] Step 4: Based on the equality condition of InSAR observations of the same-name points, a polynomial model is established for the difference in InSAR observations between the same-name points in the overlapping area.

[0012] Step 5: Determine the polynomial model weights based on InSAR coherence and solve the polynomial model parameters using the least squares method;

[0013] Step 6: Use the polynomial model to correct the InSAR observations of each platform or track one by one;

[0014] Step 7: Based on the InSAR observations corrected in step 6, data fusion is performed in the overlapping area to obtain the overall spliced InSAR observations.

[0015] Compared with the prior art, the present invention has the following significant advantages:

[0016] The method of the present invention selects points with the same name in the overlapping area and establishes a polynomial model based on the differences in InSAR observations between the points with the same name, determines the weight of the polynomial model by InSAR coherence, and solves the multi-platform, multi-orbit polynomial model parameters using the least squares method, thereby solving the problem of inconsistent multi-orbit, multi-platform benchmarks, and compensating for the problems of inconsistent observations and insufficient reliability in overlapping areas due to multiple errors in a large spatial range, thereby realizing large-scale InSAR earth observation with high temporal and spatial resolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention provides a flowchart of a multi-platform, multi-track InSAR observational data splicing method based on unconstrained adjustment.

[0018] Figure 2 This is a relationship diagram of multi-track and multi-platform InSAR observations, where (x, y) represents coordinates; 1, 2, ..., N represents the numbers of different platforms / tracks, and (az N ,rg N) represents the coordinates of platform / track N observations along the azimuth and range directions of the SAR image, “+” represents the same-name point, and the area filled with “+” represents the overlapping area of multi-track and multi-platform InSAR observations. DETAILED DESCRIPTION

[0019] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention.

[0020] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art in the art to which this invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined as such, will not be interpreted in an idealized or overly formal sense.

[0021] SAR satellite data from a single platform and orbit has limited spatial coverage and cannot meet the needs of large-scale observations. Therefore, InSAR observations from multiple platforms and orbits must be spliced together to obtain comprehensive, high-precision, large-scale surface information with a unified spatiotemporal reference. However, during the splicing process, various errors and differences in unwrapping reference points can lead to discrepancies between InSAR observations within overlapping regions of different platforms and orbits. This contradicts the theoretical assumption that observations from points with the same name should be equal. Therefore, adjustments should be performed during the splicing process.

[0022] like Figure 1 As shown in FIG, the present invention proposes a new multi-platform, multi-track InSAR observation splicing method based on unconstrained adjustment, the main steps of which are as follows:

[0023] Step 1: Obtain multi-platform, multi-orbit SAR satellite data for a certain observation period in the study area, perform interferometric processing and time series analysis on the corresponding SAR images, and obtain multiple InSAR observations.

[0024] The InSAR observations described in this invention include differential interferograms, cumulative deformation, average deformation rate, and interferometric coherence results. These results are obtained through interferometric measurement and time-series analysis using multi-temporal SAR imagery and digital elevation model (DEM) data of the study area. The general interferometric processing process includes SAR image registration, reference phase removal, atmospheric error correction, phase filtering, coherence estimation, geocoding, and phase unwrapping. Time-series InSAR data processing generally involves the selection of target points (such as permanent scatterers and distributed scattering targets) and the estimation of deformation rates and deformation amounts.

[0025] Since the spatial coverage of SAR satellite data from a single platform and a single orbit is limited, the InSAR observations obtained from the above data processing are spatially dispersed and discontinuous (e.g. Figure 2 The 1, 2, ..., N in the figure represent platform / orbit numbers), which need to be spliced to obtain large-scale InSAR observations.

[0026] Due to various factors such as errors and differences in unwrapping reference points, InSAR observations within overlapping regions of different platforms and orbits differ. This contradicts the theoretical assumption that observations from points with the same name should be equal. Therefore, adjustments must be performed on the InSAR observations when stitching them together.

[0027] Step 2: Preprocess the multiple InSAR observations obtained in step 1, including establishing a unified coordinate system, time and space resampling, and direction conversion of deformation observations.

[0028] Because the InSAR observations obtained in step 1 may originate from satellite data from different platforms and orbits, and different processing parameters and strategies are used during data processing, these InSAR observations must first be preprocessed to ensure information consistency. Specifically, a unified multi-platform, multi-orbit InSAR observation coordinate system is established to ensure spatial information consistency; resampling operations are used to ensure consistency in the temporal and spatial resolution of InSAR observations from different sources; and two-dimensional decomposition or projection transformations are used to ensure directional consistency of deformation observations.

[0029] Step 3: Determine the overlapping area of InSAR observations through spatial intersection operation and select points with the same name using InSAR coherence.

[0030] Based on the geographic information contained in the preprocessed InSAR observations, a spatial intersection operation is performed based on the geometric distribution to determine the overlapping area. Within the overlapping area, the interferometric coherence is used to select the same-name points, and the multi-platform, multi-orbit InSAR observation data at these same-name points are obtained.

[0031] Step 4: Based on the equality condition of InSAR observations of the same-name points, a polynomial model is established for the difference in InSAR observations between the same-name points in the overlapping area.

[0032] Image overlap areas (such as Figure 2 In the interferograms of different platforms and orbits, the InSAR observations of the same-name points should be theoretically equal, as shown in the figure below:

[0033] (1)

[0034] Where: is the theoretical value of the observed quantity, N is the platform / orbit number, and i represents the number of the point with the same name in the overlapping area between platform / orbit N-1 and platform / orbit N.

[0035] Due to factors such as various errors and differences in unwrapping reference points, there are certain differences between the observed quantities of the same-name points in actual observations. A polynomial model is used to model them:

[0036] (2)

[0037] Where: are the coefficients of the polynomial model to be solved for platform / track N; ( ) represents the coordinates of the InSAR observation point i on platform / track N; It represents the correction of the InSAR observation at the same name point i on platform / track N.

[0038] In actual observations, the relationship between the observation quantities of the same-name points in the overlapping area of multiple platforms and multiple orbits is:

[0039] (3)

[0040] in, is the observed value of the InSAR observation at the same-name point i in the overlapping area of platform / track N, is the observation value of the InSAR observation at the same-name point i in the overlapping area of platform / track N-1.

[0041] Step 5: Determine the polynomial model weights based on InSAR coherence and solve the polynomial model parameters using the least squares method.

[0042] The observations of the same-name points in the overlapping area of multiple platforms and multiple tracks are included in a unified process, and the adjustment observation equation is constructed according to formulas (2) and (3):

[0043] ,

[0044] Where, represents the residuals of the observations of m points with the same name in the overlapping area of platform / track N-1 and platform / track N, where N-1 and N in the superscript represent the overlapping area of platform / track N-1 and N; represents the residual of the observation of the same-name point i in the overlapping area of platform / track N-1 and platform / track N, i = 1, 2, …, m; represents the adjustment coefficient of m points of the same name in the overlapping area of platform / track N-1 and platform / track N on platform / track N-1. represents the adjustment coefficient of m points of the same name in the overlapping area of platform / track N-1 and platform / track N on platform / track N. represents the coordinates of the point i on platform / track N-1 in the overlapping area of platform / track N, represents the coordinates of the same-name point i on platform / track N in the overlapping area of platform / track N-1 and platform / track N, i = 1, 2, …, m; is the polynomial model coefficient matrix of platform / track N observations, is the difference between the observations of m points of the same name in the overlapping area of platform / track N-1 and platform / track N; and are the InSAR observation values of the same-name point i in the overlapping area of platforms / tracks N-1 and N at platforms / tracks N-1 and N-1, respectively.

[0045] Simplify formula (4) to form, in which , , , .

[0046] At this time, the least squares method can be used to solve:

[0047] (5)

[0048] Where: represents the weight matrix determined by InSAR coherence, is the weight matrix of m points with the same name in the overlapping area of platform / track N-1 and platform / track N, is the coherence of the same-name point i on platform / track N-1 in the overlapping region of platform / track N, is the coherence of the same-name point i on platform / track N in the overlapping region of platform / track N-1 and platform / track N.

[0049] Step 6: Use the polynomial model to correct the InSAR observations of each platform or track one by one.

[0050] According to the polynomial coefficients of N platforms / tracks solved in step 5 , calculate the correction value corresponding to each pixel j of the platform / track N observation according to formula (2) in step 4 , and then the original observations Correction is performed to obtain the corrected InSAR observations :

[0051] (6)

[0052] Where, is the observed value of effective pixel j in the InSAR observation of platform / track N after correction, j is the effective pixel number in the InSAR observation of platform / track N, It represents the correction value of effective pixel j in the InSAR observation of platform / track N.

[0053] Step 7: Data fusion to obtain the overall InSAR observations after splicing.

[0054] Based on the corrected InSAR observations from step 6, data fusion is performed in overlapping regions, using methods such as mean or median calculations, to ultimately obtain a stitched, integrated InSAR observation. After stitching is complete, a visual inspection is performed to verify stitching quality and smoothness, ensuring consistency across multiple platforms and orbits. Finally, the stitched image is exported to the desired format for further applications such as time series analysis and deformation interpretation.

[0055] It is particularly noted that the processes involved in the present invention, such as InSAR registration, phase unwrapping, resampling, and time series processing, are common knowledge to those skilled in the art, and therefore such basic knowledge will not be further introduced.

[0056] Based on the same technical solution, the present invention also provides an electronic device, comprising:

[0057] memory for storing computer programs;

[0058] A processor is used to implement the steps of the above-mentioned multi-platform, multi-orbit InSAR observational quantity splicing method when executing the computer program.

[0059] Based on the same technical solution, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the multi-platform, multi-orbit InSAR observational data stitching method. The computer-readable storage medium may include any medium capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0060] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0061] The above description is only a preferred embodiment of the present invention. It should be pointed out that, on this basis, various modifications and variations that can be made by technicians in the technical field to which the present invention belongs without any creative work are still within the scope of protection of the present invention.

Claims

1. A multi-platform, multi-track InSAR observation splicing method based on unconstrained adjustment, characterized in that: The specific steps of this method are as follows: Step 1: Obtain multi-platform, multi-orbit SAR satellite data for a set observation period in the study area, perform interferometric processing and time series analysis, and obtain multiple InSAR observations; Step 2: Preprocess the multiple InSAR observations obtained in step 1; Step 3: Based on the preprocessing results of step 2, determine the overlapping area of InSAR observations and select the same-name points using InSAR coherence; Step 4: Based on the equality condition of InSAR observations of the same-name points, a polynomial model is established for the difference in InSAR observations between the same-name points in the overlapping area. Step 5: Determine the polynomial model weights based on InSAR coherence and solve the polynomial model parameters using the least squares method; Step 6: Use the polynomial model to correct the InSAR observations of each platform or track one by one; Step 7: Based on the InSAR observations corrected in step 6, data fusion is performed in the overlapping area to obtain the overall spliced InSAR observations.

2. The method according to claim 1, characterized in that The interferometric processing process in step one includes SAR image registration, reference phase removal, atmospheric error correction, phase filtering, coherence estimation, geocoding, and phase unwrapping. The time series analysis processing includes target point selection, deformation rate, and deformation amount estimation.

3. The method according to claim 1, characterized in that The preprocessing in step 2 includes: Establish a unified multi-platform, multi-orbit InSAR observation coordinate system to ensure the consistency of spatial information; Through resampling operations, the consistency of temporal and spatial resolution of InSAR observations from different sources is ensured; The consistency of the deformation observation direction is ensured through two-dimensional decomposition or projection transformation.

4. The method according to claim 1, wherein In step 3, the spatial intersection operation is used to determine the overlapping area of InSAR observations: Based on the geographic information contained in the preprocessing results of step 2, a spatial intersection operation is performed according to the geometric distribution to determine the overlapping area.

5. The method according to claim 1, wherein In step 4, a polynomial model is established for the difference in InSAR observations between points with the same name in the overlapping area, specifically: ， Where N is the platform / track number, i is the point number with the same name, represents the correction of the InSAR observation of the same name point i on platform / track N, are the coefficients of the polynomial model for platform / track N, ( ) represents the coordinates of the InSAR observation point i on platform / track N.

6. The method according to claim 5, characterized in that Step 5 is as follows: Construct the adjustment observation equation: ， Where, represents the residuals of the observations of m points with the same name in the overlapping area of platform / track N-1 and platform / track N, where N-1 and N in the superscript represent the overlapping area of platform / track N-1 and N; represents the residual of the observation of the same-name point i in the overlapping area of platform / track N-1 and platform / track N, i = 1, 2, …, m; represents the adjustment coefficient of m points of the same name in the overlapping area of platform / track N-1 and platform / track N on platform / track N-1. represents the adjustment coefficient of m points of the same name in the overlapping area of platform / track N-1 and platform / track N on platform / track N. represents the coordinates of the point i on platform / track N-1 in the overlapping area of platform / track N, represents the coordinates of the same-name point i on platform / track N in the overlapping area of platform / track N-1 and platform / track N, i = 1, 2, …, m; is the polynomial model coefficient matrix of platform / track N observations, is the difference between the observations of m points of the same name in the overlapping area of platform / track N-1 and platform / track N; and are the InSAR observation values of the same-name point i in the overlapping area of platforms / tracks N-1 and N, respectively; The adjustment observation equation is simplified to The least squares method is used to solve the polynomial model coefficients; where , , , .

7. The method according to claim 6, characterized in that The least squares method is used to solve the polynomial model coefficients, specifically: ， Where, represents the weight matrix determined by InSAR coherence, is the weight matrix of m points with the same name in the overlapping area of platform / track N-1 and platform / track N, is the coherence of the same-name point i on platform / track N-1 in the overlapping region of platform / track N, is the coherence of the same-name point i on platform / track N in the overlapping region of platform / track N-1 and platform / track N.

8. The method according to claim 5, characterized in that The corrected result in step 6 is: ， Where, is the observed value of effective pixel j in the InSAR observation of platform / track N after correction, j is the effective pixel number in the InSAR observation of platform / track N, It represents the correction value of effective pixel j in the InSAR observation of platform / track N.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which implements the steps of the method according to any one of claims 1 to 8 when executed by a processor.

10. An electronic device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the method according to any one of claims 1 to 8 when executing the computer program.

Images