Time sequence InSAR ionized layer error robust correction method based on Re-RSSI

Abstract

The invention discloses a time sequence InSAR ionosphere error robust correction method based on Re-RSSI, and the method comprises the steps: estimating a space-time baseline of each SAR image pair, determining an SAR image pair set needed by time sequence solution according to a preset space-time baseline threshold value, carrying out the band-pass filtering of an original SLC image, carrying out the registration of the full-bandwidth and sub-band SLC image set, and carrying out the interference processing. The method comprises the following steps: obtaining a full-bandwidth and high-low-frequency double-difference interferogram, carrying out phase unwrapping, obtaining a full-bandwidth and high-low-frequency double-difference unwrapping phase, carrying out filtering processing, obtaining a filtering unwrapping interferogram, estimating the full-bandwidth and high-low-frequency double-difference time sequence ionosphere phase in combination with a weighted least square algorithm, and removing an estimation result from an original InSAR (Interferometric Synthetic Aperture Radar). And ionization error correction is completed. According to the method, the ionosphere error correction capability is effectively improved, the influence of the ionosphere error on the monitoring precision is reduced, and the high precision of the geological disaster monitoring result is ensured.

Description

Technical Field

[0001] The present invention relates to the field of InSAR ionospheric error correction, and in particular to a time-series InSAR ionospheric error robust correction method based on Re-RSSI. Background Art

[0002] Low-frequency SAR imagery, due to its long wavelength and strong penetration, offers significant advantages in geological disaster monitoring in complex terrain and vegetation-covered areas. However, its measurement accuracy is severely affected by ionospheric interference. Currently, ionospheric error correction methods for low-frequency TS-InSAR mainly rely on RSSI and Re-RSSI methods.

[0003] Limited by the limited range bandwidth of current SAR imagery, the RSSI method is generally less robust and tends to amplify decorrelation noise in the interferogram, reducing its sensitivity to ionospheric errors. Furthermore, during time-series network construction, the ionospheric error correction performance of RSSI-based TS-InSAR is significantly degraded by differences in noise levels between different interferometers and phase unwrapping errors. Although the Re-RSSI method significantly improves its robustness compared to the RSSI method, the linear parameters of the high- and low-frequency InSAR double-difference phase remain large when estimating the ionospheric error phase, resulting in a significant noise impact.

[0004] This paper provides a robust ionospheric correction method for time-series InSAR based on Re-RSSI. By combining differential time series with time-series synthetic aperture radar interferometry and applying reconstructed range-spectral spectral analysis to correct ionospheric errors, the accuracy and robustness of ionospheric phase estimation are significantly improved. Summary of the Invention

[0005] The purpose of the present invention is to provide a time-series InSAR ionospheric error robust correction method based on Re-RSSI.

[0006] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0007] The present invention provides a time-series InSAR ionospheric error robust correction method based on Re-RSSI, comprising the following steps:

[0008] Step S1, estimating the spatiotemporal baseline of each synthetic aperture radar image pair, determining a set of synthetic aperture radar image pairs required for time series solution according to a preset spatiotemporal baseline threshold, and performing bandpass filtering on the single-look complex images in the synthetic aperture radar image pair set to obtain a full-bandwidth and sub-band single-look complex image set;

[0009] Step S2, performing interference processing on the full-bandwidth and sub-band single-view complex image sets to obtain a full-bandwidth interferogram and a high- and low-frequency double-difference interferogram, and further performing phase unwrapping to obtain an unwrapped phase of the full-bandwidth and high- and low-frequency double-differences;

[0010] Step S3, performing spatial filtering on the unwrapped phase of the full bandwidth and high- and low-frequency double-difference, and estimating the time-series ionospheric phase based on the filtering result;

[0011] Step S4: removing the time series ionospheric phase from the original time series phase to obtain an error-corrected time series deformation phase.

[0012] Preferably, the method of estimating the spatiotemporal baseline of each SAR image pair in step S1 is specifically: estimating the spatiotemporal baseline of each SAR image pair using the date and observation geometry of each SAR image.

[0013] Preferably, the method of performing bandpass filtering on the single-view complex images in the synthetic aperture radar image pair set in step S1 is specifically as follows:

[0014] Bandpass filtering is performed on the single-view complex image to obtain a single-view complex image set with full bandwidth and low-frequency and high-frequency sub-bands;

[0015] The single-look complex image sets of full bandwidth and low-frequency and high-frequency subbands are registered.

[0016] Preferably, the method of performing interference processing on the full-bandwidth and sub-band single-view complex image sets in step S2 is specifically: performing full-bandwidth interference processing on the full-bandwidth single-view complex image set, and performing double differential interference processing on the low-frequency and high-frequency sub-band single-view complex image sets.

[0017] Preferably, the method of performing spatial filtering on the unwrapped phase of the full bandwidth and high- and low-frequency double-difference in step S3 is specifically: removing abnormal data by median filtering, and reducing high-frequency noise in the image by Gaussian filtering.

[0018] Preferably, the filtering result in step S3 includes the filtered full bandwidth and high- and low-frequency double-differential interference phase sets.

[0019] Preferably, the method steps for estimating the time series ionospheric phase based on the filtering result in step S3 are specifically:

[0020] The full bandwidth and double-difference timing phase are estimated by weighted least squares method, and the expression is:

[0021]

[0022] Where, is the estimated value of the timing phase, A is the design matrix, T represents the transpose operation, W is the M×M weight matrix, M is the number of interference pairs, is the unwrapped phase variation of the interference pair;

[0023] The time series ionospheric phase is estimated based on the full bandwidth and double difference time series phase, and the expression is:

[0024]

[0025] Where, is the temporal ionospheric phase, f L and f H are the center frequencies of the low-frequency and high-frequency sub-bands respectively, f0 is the radar carrier frequency, and They are full bandwidth timing phase estimation value and double differential timing phase estimation value respectively;

[0026] in,

[0027] f L =f0-Δf / 2

[0028] f H =f0+Δf / 2

[0029] Δf=(1-k)·B rc

[0030] Where Δf is the center frequency difference between the high-frequency and low-frequency sub-bands, k is the scaling factor, and B rc is the range bandwidth of the synthetic aperture radar system.

[0031] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0032] (1) The ionospheric error correction in the present invention combines the reconstructed range-direction spectrum method and the weighted least squares method to linearly combine the double difference and full-bandwidth time series phase, which can accurately estimate and correct the ionospheric phase error. Compared with the traditional ionospheric error correction technology, it can significantly reduce the interference of the phase unwrapping process and noise differences on the accuracy of the time series ionospheric phase estimation, and improve the accuracy of geological disaster monitoring results;

[0033] (2) The present invention not only reduces the influence of various types of noise through spatial filtering processing, but also significantly improves the robustness of time-series ionospheric phase error estimation and correction, thereby enhancing the reliability of geological disaster monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a robust correction method for time-series InSAR ionospheric errors based on Re-RSSI of the present invention;

[0035] Figure 2 Schematic diagram of a flow chart in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0037] Reference Figure 1 、 Figure 2 As shown, the present invention provides a time-series InSAR ionospheric error robust correction method based on Re-RSSI, comprising:

[0038] Step S1, estimating the spatiotemporal baseline of each synthetic aperture radar image pair, specifically: estimating the spatiotemporal baseline of each synthetic aperture radar image pair based on the date and observation geometry of each synthetic aperture radar image; determining a set of synthetic aperture radar image pairs required for time series solution according to a preset spatiotemporal baseline threshold, performing bandpass filtering on the single-look complex images in the set of each synthetic aperture radar image pair, and then performing registration, specifically: performing bandpass filtering on the single-look complex images to obtain a set of single-look complex images of full bandwidth, low frequency, and high frequency subbands with different center frequencies; performing coarse registration on the single-look complex image sets of full bandwidth, low frequency, and high frequency subbands based on observation geometry parameters to obtain a coarse registration polynomial, resampling the auxiliary image to align it with the reference system of the primary image, refining the coarse registration offset polynomial based on DEM data, and using it to perform fine registration to complete the registration process; obtaining a registered single-look complex image set of full bandwidth and subbands.

[0039] Step S2, performing interference processing on the full-bandwidth and sub-band single-view complex image sets, specifically: performing full-bandwidth interference processing on the full-bandwidth single-view complex image set, and performing double-differential interference processing on the sub-band single-view complex image set; obtaining a full-bandwidth interference diagram and a high- and low-frequency double-differential interference diagram, and further performing phase unwrapping to obtain an unwrapped phase of the full-bandwidth and high- and low-frequency double-differentials.

[0040] Step S3: spatially filter the unwrapped phases of the full bandwidth and high- and low-frequency double-differences. Specifically, the following steps are performed: removing outliers in the double-difference interferometry phase by median filtering, and reducing high-frequency noise in the full bandwidth and sub-band unwrapped interferometry phases by Gaussian filtering; obtaining an estimated data set including the full bandwidth phase, high- and low-frequency double-difference phases, and coherence data after spatial filtering; and performing estimation based on the full bandwidth and high- and low-frequency double-difference phases. Specifically, the following steps are performed:

[0041] The full bandwidth and double-difference timing phase are estimated by weighted least squares method, and the expression is:

[0042]

[0043] Where, is the estimated value of the timing phase, A is the design matrix, T represents the transpose operation, W is the M×M weight matrix, M is the number of interference pairs, is the unwrapped phase variation of the interference pair;

[0044] The time series ionospheric phase is estimated based on the full bandwidth and double difference time series phase, and the expression is:

[0045]

[0046] Where, is the temporal ionospheric phase, f L and f H are the center frequencies of the low-frequency and high-frequency sub-bands respectively, f0 is the radar carrier frequency, and They are full bandwidth timing phase estimation value and double differential timing phase estimation value respectively;

[0047] in,

[0048] f L =f0-Δf / 2

[0049] f H =f0+Δf / 2

[0050] Δf=(1-k)·B rc

[0051] Where Δf is the center frequency difference between the high-frequency and low-frequency sub-bands, k is the scaling factor, and B rc is the range bandwidth of the synthetic aperture radar system;

[0052] It should be explained that the spatial filtering method is used for filtering processing, which minimizes the interference of noise on subsequent data processing by performing median and Gaussian filtering on the phase image or data. The purpose is to remove outliers and noise and enhance the smoothness of the signal.

[0053] It should be explained that, considering the use of the same filtering method and filtering parameters in the process of time-series InSAR data processing, there are differences in the noise levels in the ionospheric phases of different full-bandwidth InSAR interferometer pairs estimated using the Re-RSSI method, which affects the estimation quality of the ionospheric phase time series and reduces the correction accuracy of the TS-InSAR large-scale ionospheric error. In order to suppress the impact of this noise difference and take into account the influence of the noise level differences between the full-bandwidth differential InSAR phases, the weighted least squares algorithm is used to estimate the time series phase.

[0054] Step S4: removing the time series ionospheric phase from the original time series phase to obtain an error-corrected time series deformation phase.

[0055] The full Chinese name of Re-RSSI is Reconstructed Range Spectral Interferometry, the Chinese name of InSAR is Synthetic Aperture Radar Interferometry, the Chinese name of DInSAR is Differential Synthetic Aperture Radar Interferometry, the Chinese name of TS-InSAR is Time Series Synthetic Aperture Radar Interferometry, SLC image is single-look complex image, and SAR image is synthetic aperture radar image.

[0056] The above content is merely an example and explanation of the structure of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A robust correction method for ionospheric errors in time series InSAR based on Re-RSSI, characterized in that: The following steps are involved: Step S1, estimating the spatiotemporal baseline of each synthetic aperture radar image pair, determining a set of synthetic aperture radar image pairs required for time series solution according to a preset spatiotemporal baseline threshold, and performing bandpass filtering on the single-look complex images in the synthetic aperture radar image pair set to obtain a full-bandwidth and sub-band single-look complex image set; Step S2, performing interference processing on the full-bandwidth and sub-band single-view complex image sets to obtain a full-bandwidth interferogram and a high- and low-frequency double-difference interferogram, and further performing phase unwrapping to obtain an unwrapped phase of the full-bandwidth and high- and low-frequency double-differences; Step S3, performing spatial filtering on the unwrapped phase of the full bandwidth and high- and low-frequency double-difference, and estimating the time-series ionospheric phase based on the filtering result; Step S4: removing the time series ionospheric phase from the original time series phase to obtain an error-corrected time series deformation phase.

2. The method for robust correction of ionospheric errors in time series InSAR based on Re-RSSI according to claim 1, characterized in that: The method for estimating the spatiotemporal baseline of each SAR image pair in step S1 is specifically: estimating the spatiotemporal baseline of each SAR image pair using the date and observation geometry of each SAR image.

3. The robust correction method for time-series InSAR ionospheric error based on Re-RSSI according to claim 1, characterized in that: The method of performing bandpass filtering on the single-view complex image in the synthetic aperture radar image pair set in step S1 is specifically as follows: Bandpass filtering is performed on the single-view complex image to obtain a single-view complex image set with full bandwidth and low-frequency and high-frequency sub-bands; The single-look complex image sets of full bandwidth and low-frequency and high-frequency subbands are registered.

4. The method for robust correction of ionospheric errors in time series InSAR based on Re-RSSI according to claim 1, characterized in that: The method of performing interference processing on the full-bandwidth and sub-band single-view complex image sets in step S2 is specifically: performing full-bandwidth interference processing on the full-bandwidth single-view complex image set, and performing double differential interference processing on the low-frequency and high-frequency sub-band single-view complex image sets.

5. The method for robust correction of ionospheric errors of time series InSAR based on Re-RSSI according to claim 1, characterized in that: The method of performing spatial filtering on the unwrapped phase of the full bandwidth and high- and low-frequency double-difference in step S3 is specifically: removing abnormal data by median filtering, and reducing high-frequency noise in the image by Gaussian filtering.

6. The method for robust correction of ionospheric errors of time series InSAR based on Re-RSSI according to claim 1, characterized in that: The filtering result in step S3 includes the filtered full bandwidth and high- and low-frequency double-differential interference phase sets.

7. The method for robust correction of ionospheric errors of time series InSAR based on Re-RSSI according to claim 1, characterized in that: The method steps for estimating the time series ionospheric phase based on the filtering results in step S3 are specifically as follows: The full bandwidth and double-difference timing phase are estimated by weighted least squares method, and the expression is: Where, is the estimated value of the timing phase, A is the design matrix, T represents the transpose operation, W is the M×M weight matrix, M is the number of interference pairs, is the unwrapped phase variation of the interference pair; The time series ionospheric phase is estimated based on the full bandwidth and double difference time series phase, and the expression is: Where, is the temporal ionospheric phase, f L and f H are the center frequencies of the low-frequency and high-frequency sub-bands respectively, f0 is the radar carrier frequency, and They are full bandwidth timing phase estimation value and double differential timing phase estimation value respectively; in, f L =f0-Δf / 2 f H =f0+Δf / 2 Δf=(1-k)·B rc Where Δf is the center frequency difference between the high-frequency and low-frequency sub-bands, k is the scaling factor, and B rc is the range bandwidth of the synthetic aperture radar system.

Images