A multi-baseline insar residual phase error correction method and system based on flat prior information

Abstract

The application provides a multi-baseline InSAR residual phase error correction method and system based on flat prior information, selects two flat regions with different sizes through the topographic prior knowledge of a research region, constructs a residual linear phase model, removes the residual linear error phase of the multi-baseline InSAR interference phase, and realizes the functions of improving the multi-baseline InSAR phase error accuracy and correction speed.The application improves the correction accuracy of the multi-baseline InSAR interference phase after flat removal, has high phase error correction accuracy in the small baseline case, has high multi-baseline InSAR terrain reconstruction accuracy, and provides a reliable data basis for the multi-baseline InSAR terrain accurate reconstruction.The application has fast correction error phase process speed, reduces the consumption of calculation resources and the hardware cost of the system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of interferometric synthetic aperture radar data processing, and particularly relates to a multi-baseline InSAR residual phase error correction method and system based on flat ground prior information. BACKGROUND

[0002] Interferometric Synthetic Aperture Radar (InSAR) is an image geodetic surveying technology developed rapidly in recent years, which uses SAR image data observed from different angles to survey the ground surface topography. Multi-baseline InSAR is to combine SAR image data observed from different angles to form a virtual aperture in the height direction, so as to realize three-dimensional reconstruction of the ground surface topography and greatly improve the topographic elevation measurement accuracy. Therefore, multi-baseline InSAR has become a reliable means for current InSAR topographic surveying, such as the multi-baseline InSAR system formed by the four satellites of Hongmu-1, which realizes accurate three-dimensional reconstruction of the global complex ground surface topography.

[0003] However, in the data acquisition process of multi-baseline InSAR, it is usually affected by deterministic errors caused by InSAR system errors including baseline error, calibration error, atmospheric delay and other factors. If these deterministic errors are not eliminated, they will seriously restrict the height measurement accuracy of multi-baseline InSAR and even the success or failure. Therefore, phase error correction is one of the key technologies of multi-baseline InSAR.

[0004] There are a large number of phase error correction methods based on coherence measure criteria, which calculate residual error phase along the multi-baseline dimension pixel by pixel, update the terrain reconstruction information at the same time, and alternately iterate to optimize the residual error phase and terrain height. This kind of method requires a large number of interferograms and accurate residual error phase model assumptions, and the residual phase error correction speed is slow. Moreover, when the number of interferograms is small and the target signal-to-noise ratio is low, the residual phase error correction accuracy is low.

[0005] The phase error of InSAR caused by system error, atmospheric delay and other factors usually shows a linear relationship along the range and azimuth directions, and the true interferometric phase of flat ground is usually close to zero. Therefore, it is urgent to propose a new method for multi-baseline InSAR residual phase error correction to improve the phase error correction speed and the phase error correction accuracy under small baseline conditions. SUMMARY

[0006] The technical problem to be solved by the present application is to provide a multi-baseline InSAR residual phase error correction method and system based on flat ground prior information, which is used to improve the accuracy and correction speed of multi-baseline InSAR phase error.

[0007] The application is a kind of multi-baseline InSAR residual phase error correction method based on flat prior information, which comprises the following steps:

[0008] S1: Obtain multi-baseline InSAR images and perform interference preprocessing to obtain multi-baseline InSAR interference phases after flat removal;

[0009] S2: Remove the residual linear error phase of the multi-baseline InSAR interference phase after flat removal to obtain the multi-baseline InSAR interference phase after correction of the phase error; the specific steps include:

[0010] S21: Select a flat area and a window flat area in the flat area based on prior knowledge;

[0011] S22: Remove the main part of the residual linear error phase of the multi-baseline InSAR interference phase after flat removal in the window flat area;

[0012] S23: Remove the remaining part of the residual linear error phase of the multi-baseline InSAR interference phase after flat removal in the flat area;

[0013] S3: Reconstruct the multi-baseline InSAR terrain or three-dimensional reconstruction tomographic SAR point cloud based on the multi-baseline InSAR interference phase after correction of the phase error.

[0014] According to the above scheme, in step S1, the interference preprocessing includes selecting a main image, multi-baseline SAR image registration, generating an interference phase, and performing flat removal on the interference phase.

[0015] According to the above scheme, in step S2, the residual linear error phase includes InSAR system error, baseline error and atmospheric delay.

[0016] According to the above scheme, in step S21, the specific steps are:

[0017] According to the prior knowledge of the terrain of the research area, the same flat area is circled in the multi-baseline InSAR interference phase, and the same window flat area is selected from the flat area.

[0018] Further, in step S22, the specific steps are:

[0019] S221: Construct a residual linear phase model to fit the parameters of the residual linear phase in the multi-baseline InSAR interference phase after flat removal in the window flat area;

[0020] S222: Substitute the parameters of the residual linear phase in the window flat area into the residual linear phase model to solve the residual linear phase of all pixels of the multi-baseline InSAR interference phase.

[0021] S223: subtract the residual linear phase obtained in step S222 from the de-flattened multi-baseline InSAR interferometric phase to remove the main part of the residual linear phase in the de-flattened multi-baseline InSAR interferometric phase, and obtain a multi-baseline InSAR interferometric phase with a remaining part of the residual linear phase.

[0022] Further, in step S23, the specific steps are as follows:

[0023] S231: fitting the parameters of the residual linear phase in the de-flattened multi-baseline InSAR interferometric phase in a flat area;

[0024] S232: substituting the parameters of the residual linear phase fitted in the flat area into the residual linear phase model to solve the residual linear phase of all pixels of the multi-baseline InSAR interferometric phase with the remaining part of the residual linear phase;

[0025] S233: subtracting the residual linear phase obtained in step S232 from the de-flattened multi-baseline InSAR interferometric phase to remove the error of the multi-baseline InSAR interferometric phase with the remaining part of the residual linear phase, and obtain the multi-baseline InSAR interferometric phase after correcting the phase error.

[0026] According to the above scheme, in step S22, the parameters of the fitted residual linear phase include a coefficient of a first-order term along the range direction, a coefficient of a first-order term along the azimuth direction, and a constant term parameter.

[0027] A multi-baseline InSAR residual phase error correction system based on flat prior information includes a preprocessing module, a phase error correction module, and a terrain reconstruction module. The preprocessing module is used to obtain multi-baseline InSAR images and perform interference preprocessing to obtain de-flattened multi-baseline InSAR interferometric phase. The phase error correction module is used to remove the residual linear error phase of the de-flattened multi-baseline InSAR interferometric phase to obtain the multi-baseline InSAR interferometric phase after correcting the phase error. The terrain reconstruction module is used to reconstruct the multi-baseline InSAR terrain or three-dimensional reconstruction tomographic SAR point cloud based on the multi-baseline InSAR interferometric phase after correcting the phase error.

[0028] Further, the phase error correction module comprises a flat area selection module, a main error phase component removal module and a residual error phase component removal module; the flat area selection module is used for selecting flat areas and window flat areas in the flat areas based on prior knowledge; the main error phase component removal module is used for removing main components of residual linear error phases of the multi-baseline InSAR interference phase after flattening in the window flat areas; and the residual error phase component removal module is used for removing residual components of the residual linear error phases of the multi-baseline InSAR interference phase after flattening in the flat areas.

[0029] Further, the application further comprises a memory, which stores a computer program executable by a computer processor, and the computer program executes a multi-baseline InSAR residual phase error correction method based on flat area prior information.

[0030] The application has the following advantages:

[0031] 1. The application is a multi-baseline InSAR residual phase error correction method and system based on flat area prior information, which selects flat areas of different sizes based on the topographic prior knowledge of a study area, constructs a residual linear phase model, removes residual linear error phases of the multi-baseline InSAR interference phase with fewer maps and correction times, and realizes the functions of improving the multi-baseline InSAR phase error accuracy and correction speed.

[0032] 2. The application improves the correction accuracy of the multi-baseline InSAR interference phase after flattening, has high phase error correction accuracy in the small baseline case, has high multi-baseline InSAR terrain reconstruction accuracy, and provides a reliable data basis for multi-baseline InSAR terrain accurate reconstruction.

[0033] 3. The application has fast correction error phase process, reduces the consumption of computing resources and the hardware cost of the system. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 is a flowchart of an embodiment of the application.

[0035] Figure 2 is any one of the interference phase maps generated by simulation of an embodiment of the application.

[0036] Figure 3 is an interference phase map after flattening of an embodiment of the application.

[0037] Figure 4 is an interference phase map in which most residual linear errors are removed of an embodiment of the application.

[0038] Figure 5 is an interference phase map in which residual small linear errors are removed of an embodiment of the application.

[0039] Figure 6 is a phase-corrected error post multi-baseline InSAR reconstruction topographic result map of an embodiment of the present application. DETAILED DESCRIPTION

[0040] The application will be further described below in conjunction with the drawings and specific embodiments.

[0041] Embodiment 1

[0042] Referring to Figure 1 , the embodiment includes the following steps:

[0043] Step 1: Obtain SAR data collected by multiple antennas or multiple passes of a radar and its ephemeris parameters; perform interference preprocessing on the multi-baseline SAR image, including selecting a main image, registering the multi-baseline SAR image, generating interference phase, and removing flat ground interference phase from the interference phase, thereby obtaining the multi-baseline InSAR interference phase after flat ground removal;

[0044] Step 2: Remove residual linear error phase in the multi-baseline InSAR interference phase after flat ground removal due to InSAR baseline error, baseline error, and atmospheric delay, and the specific steps are as follows:

[0045] Step 2.1: According to the prior knowledge of the terrain of the study area, circumscribe a large and wide flat area R2 in the multi-baseline InSAR interference phase, and select a small flat area R1 from it;

[0046] Step 2.2: Construct the following residual linear phase model

[0047] (1)

[0048] wherein is the ideal residual phase error due to baseline error, interference scaling error, and atmospheric delay error, , are the distance and azimuth indexes, respectively, , is a first-order term coefficient, is a constant term, is a random noise phase, is the interference phase after flat ground removal.

[0049] Based on equation (1), extrapolate to any single baseline InSAR interference phase after flat ground removal, and combine the interference phases of all pixels in the R1 region to obtain the interference phase equation after flat ground removal

[0050] (2)

[0051] Solve the parameters of the most residual linear phase in R1 area, i.e. the first term coefficient along the range direction

[0052] (3)

[0053] Solve the parameters of the most residual linear phase in R1 area, i.e. the first term coefficient along the range direction , the first term coefficient along the azimuth direction and the constant term parameter .

[0054] Step 2.3: Substitute the parameters solved in step 2.2 , , into all pixels to obtain the fitted residual interference phase error under the baseline condition, i.e.

[0055] (4)

[0056] wherein is the fitted residual interference phase of all pixels of the baseline interference phase; subtract the fitted residual interference phase from the leveled interference phase to obtain the interference phase with the most residual linear phase removed under the baseline condition; similarly, extrapolate to all multi-baseline InSAR interference phase conditions to obtain the multi-baseline InSAR interference phase with the most residual linear phase removed.

[0057] Step 2.4: For the small part of residual linear phase remaining in any multi-baseline InSAR interference phase in step 2.3, extrapolate steps 2.2 and 2.3 to step 2.4.

[0058] First, as in step 2.2, the interference phase equations of all pixels in R2 area are solved to obtain the leveled interference phase, and the parameters of the small part of residual linear phase in R2 area are solved by using the least square method, i.e. the first term coefficient along the range direction , the first term coefficient along the azimuth direction and the constant term parameter .

[0059] Then, substitute the solved , , parameters into all pixels to obtain the fitted small residual interference phase error under the baseline condition, and accurately remove the small residual linear phase error of the multi-baseline InSAR interference phase with the most residual linear phase removed in step 2.3 to obtain the multi-baseline InSAR interference phase with residual linear error accurately corrected.

[0060] Step 3: The phase error corrected multi-baseline InSAR interferometric phase is used for multi-baseline InSAR terrain reconstruction or tomographic SAR point cloud 3D reconstruction.

[0061] It should be understood that the magnitude of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0062] Embodiment 2

[0063] In this embodiment, the steps of embodiment 1 are used to correct and simulate 20 scenes of Ku band nest building SAR data, and 19 interferometric phase maps are generated, Figure 2 for any one of the interferometric phase maps; Figure 3 for the interferometric phase map after flattening, it can be seen that the flat area stripe is effectively removed after flattening; Figure 4 for the interferometric phase map in which most of the residual linear errors are removed by step 2.3 of the present application; Figure 5 for the interferometric phase map in which residual small linear errors are removed by step 2.4 of the present application, it can be seen that after the phase error correction of the present application, the interferometric phase of the flat area is calibrated to the vicinity of zero value, indicating that the accuracy of the present application for correcting the multi-baseline InSAR interferometric phase error is high.

[0064] In addition, the time spent for correcting the residual phase error of the 19 multi-baseline interferometric phase maps by using this embodiment is not more than 1 minute. Figure 6 for the terrain result reconstructed by the multi-baseline InSAR technology, it can be seen that the flat area and the building height are effectively recovered.

[0065] The above embodiments are only used to illustrate the design idea and characteristics of the present application, and its purpose is to enable those skilled in the art to understand the content of the present application and to implement it, and the protection scope of the present application is not limited to the above embodiments. Therefore, any equivalent changes or modifications made according to the principles and design ideas disclosed by the present application are within the protection scope of the present application.

Claims

1. A multi-baseline InSAR residual phase error correction method based on flat terrain prior information, characterized by: The following steps are involved: S1: Acquire multi-baseline InSAR images and perform interferometric preprocessing to obtain the multi-baseline InSAR interferometric phase after removing the flat ground; S2: Removing the residual linear error phase of the multi-baseline InSAR interferometry phase after the flattening process to obtain the multi-baseline InSAR interferometry phase after the phase error correction. The specific steps include: S21: based on the prior knowledge of terrain and the interference phase fringe information, selecting the flat area and the window flat area in the flat area; S22: the main part of the residual linear error phase of the multi-baseline InSAR interferometric phase after removing the flat ground in the window flat ground area; S23: the remaining part of the residual linear error phase of the multi-baseline InSAR interferometric phase after removing the flat land in the flat land area; S3: Reconstruct multi-baseline InSAR terrain or 3D tomographic SAR point cloud based on the multi-baseline InSAR interferometric phase after phase error correction.

2. The multi-baseline InSAR residual phase error correction method based on flat terrain prior information according to claim 1, characterized in that: In the step S1, the interferometric preprocessing includes selecting a main image, multi-baseline SAR image registration, generating an interferometric phase, and performing ground-free interferometric phase removal on the interferometric phase.

3. The multi-baseline InSAR residual phase error correction method based on flat terrain prior information according to claim 1, characterized in that: In step S2, the residual linear error phase includes InSAR system error, baseline error and atmospheric delay.

4. The multi-baseline InSAR residual phase error correction method based on flat terrain prior information according to claim 1, characterized in that: In the step S21, the specific steps are: According to the prior knowledge of the terrain in the study area and the interferometric phase fringe information, the same flat area is delineated in the area without phase jump in the multi-baseline InSAR interferometric phase, and the same window flat area is selected from the flat area.

5. The multi-baseline InSAR residual phase error correction method based on flat terrain prior information according to claim 4, characterized in that: In the step S22, the specific steps are: S221: Construct a residual linear phase model to fit the parameters of the residual linear phase in the multi-baseline InSAR interferometric phase after removing the flat ground in the window flat ground area; S222: Substitute the parameters of the residual linear phase fitted in the flat area of ​​the window into the residual linear phase model to solve the residual linear phase of all pixels of the multi-baseline InSAR interferometric phase; S223: Subtract the residual linear phase obtained in step S222 from the multi-baseline InSAR interferometry phase after leveling to remove the main part of the residual linear phase in the multi-baseline InSAR interferometry phase, and obtain the multi-baseline InSAR interferometry phase of the remaining part of the residual linear phase.

6. The multi-baseline InSAR residual phase error correction method based on flat terrain prior information according to claim 5, characterized in that: In the step S23, the specific steps are: S231: Parameters of the residual linear phase in the multi-baseline InSAR interferometric phase after fitting the flat terrain in the flat terrain area; S232: Substituting the parameters of the residual linear phase fitted in the flat area into the residual linear phase model, and solving the residual linear phase of all pixels of the multi-baseline InSAR interferometric phase image for the remaining part of the residual linear phase; S233: Subtract the residual linear phase obtained in step S232 from the residual linear phase in the multi-baseline InSAR interferometry phase after leveling to remove the error of the multi-baseline InSAR interferometry phase of the remaining residual linear phase, and obtain the multi-baseline InSAR interferometry phase after the phase error is corrected.

7. The multi-baseline InSAR residual phase error correction method based on flat terrain prior information according to claim 1, characterized in that: In step S22, the parameters of the fitted residual linear phase include the linear term coefficient along the range direction, the linear term coefficient along the azimuth direction, and the constant term parameter.

8. A multi-baseline InSAR residual phase error correction system based on flat terrain prior information, characterized by: It includes a pre-processing module, a phase error correction module and a terrain reconstruction module; The preprocessing module is used to acquire multi-baseline InSAR images and perform interferometric preprocessing to obtain the multi-baseline InSAR interferometric phase after flattening. The phase error correction module is used to remove the residual linear error phase of the multi-baseline InSAR interferometric phase after flattening, and obtain the multi-baseline InSAR interferometric phase after phase error correction; the phase error correction module includes a flat ground area selection module, an error phase main component removal module and an error phase residual component removal module; the flat ground area selection module is used to select a flat ground area and a window flat ground area in the flat ground area based on prior knowledge; the error phase main component removal module is used to remove the main component of the residual linear error phase of the multi-baseline InSAR interferometric phase after flattening in the window flat ground area; the error phase residual component removal module is used to remove the residual component of the residual linear error phase of the multi-baseline InSAR interferometric phase after flattening in the flat ground area; The terrain reconstruction module is used to reconstruct multi-baseline InSAR terrain or three-dimensionally reconstruct tomographic SAR point cloud based on the multi-baseline InSAR interferometric phase after phase error correction.

9. The multi-baseline InSAR residual phase error correction system based on flat terrain prior information according to claim 8, characterized in that: The invention also includes a memory storing a computer program executable by a computer processor, wherein the computer program executes the multi-baseline InSAR residual phase error correction method based on flat ground prior information as claimed in any one of claims 1 to 7.

Images