High-precision deformation monitoring and target imaging method based on MIMO-SAR

Abstract

The invention relates to the technical field of radar remote sensing, and discloses a high-precision deformation monitoring and target imaging method based on MIMO-SAR. The method comprises the following steps: receiving an original MIMO-SAR echo data stream and a Beidou navigation system positioning data stream acquired by an image sensor; performing standardized preprocessing on the original echo data to generate standard radar data blocks; aligning radar data blocks and Beidou positioning data time and space, and establishing a mapping relation between data and a spatial position; performing sub-aperture division on the aligned data blocks to generate a multi-sub-aperture data set; performing beam forming processing on each sub-aperture data set to obtain a multi-view complex image; extracting an interference phase of a plurality of images and carrying out geometric correction by using Beidou data; unwrapping the corrected phase to obtain an absolute phase field; inverting a millimeter-level deformation sequence based on the absolute phase field; dynamically updating imaging focusing parameters according to the deformation sequence time evolution characteristics; and performing high-resolution imaging on subsequent data blocks by using the updated parameters, and outputting a deformation monitoring result and a target image.

Description

Technical Field

[0001] This invention relates to the field of radar remote sensing technology, specifically to a high-precision deformation monitoring and target imaging method based on MIMO-SAR. Background Technology

[0002] Current synthetic aperture radar (SAR) deformation monitoring primarily employs differential interferometry (DI), largely based on full-aperture data processing. This single-viewpoint approach makes it difficult to eliminate the effects of atmospheric delay and orbital errors. Existing technologies have limitations in deformation monitoring accuracy, rigid data processing workflows, and fail to fully utilize the complementarity of multi-view observations. Interferometric phase is susceptible to noise interference, geometric correction is insufficient, and phase unwrapping accuracy is inadequate, leading to difficulties in millimeter-level deformation inversion. Imaging parameters are typically preset and fixed, unable to be dynamically adjusted according to target deformation, making it difficult to meet the high-precision monitoring requirements of complex scenarios.

[0003] Traditional methods suffer from significant shortcomings in data utilization and adaptive processing. The spatiotemporal alignment accuracy between radar echo data and positioning information is low, resulting in large mapping errors. Sub-aperture segmentation strategies are simplistic and fail to optimize the diversity of multi-view observations. Beamforming algorithms are computationally complex and lack real-time performance. Interferometric phase correction relies on idealized models, and residual errors affect the accuracy of deformation inversion. Phase unwrapping is susceptible to terrain undulations and noise, leading to unstable absolute phase field solutions. Imaging parameter updates lag, failing to respond in real-time to changes in target state, resulting in degraded imaging quality. Summary of the Invention

[0004] The purpose of this invention is to provide a high-precision deformation monitoring and target imaging method based on MIMO-SAR to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, this invention provides a high-precision deformation monitoring and target imaging method based on MIMO-SAR, the method comprising: Receive raw MIMO-SAR echo data stream from the image sensor and positioning data stream from the BeiDou Navigation Satellite System; The original MIMO-SAR echo data stream is preprocessed using data standardization to generate radar data blocks in a standard format; The standard format radar data blocks are spatiotemporally aligned with the positioning data stream of the BeiDou navigation system to establish a mapping relationship between each radar data block and its corresponding spatial location. The spatiotemporally aligned radar data blocks are divided into sub-apertures to generate multiple sub-aperture datasets. Beamforming processing is performed on each of the sub-aperture data sets to obtain a set of complex images with different viewpoints; The interference phase information of each group of complex images is extracted, and the interference phase information is geometrically corrected using the positioning data of the Beidou navigation system; The geometrically corrected interference phase information is unwrapped to obtain the absolute phase field; Based on the absolute phase field, the millimeter-level deformation sequence of the monitored target is obtained by inversion; Based on the temporal evolution characteristics of the millimeter-level deformation sequence, the imaging focusing parameters are dynamically updated; Using the updated imaging focusing parameters, high-resolution imaging is performed on the subsequently input standard format radar data blocks, and deformation monitoring results and target images are output.

[0006] Preferably, the data normalization preprocessing of the raw MIMO-SAR echo data stream includes: The original MIMO-SAR echo data stream is parsed to separate the header information and payload data; Invalid values ​​are removed and outliers are corrected in the payload data; The amplitude of the corrected payload data is normalized to ensure that the data amplitude is within a preset dynamic range. The amplitude-normalized data is divided into segments according to a preset time window to generate radar data blocks in the standard format.

[0007] Preferably, the step of spatiotemporally aligning the standard format radar data block with the positioning data stream of the BeiDou navigation system includes: Extract timestamps and three-dimensional coordinate information from the positioning data stream of the BeiDou navigation system; Extract the data acquisition timestamp from the header information of the radar data block in the standard format; A time-indexed lookup table is established to match the data acquisition timestamp of each radar data block with the closest timestamp of the BeiDou navigation system. The three-dimensional coordinate information of the successfully matched BeiDou navigation system is assigned to the corresponding radar data block to complete the spatiotemporal alignment.

[0008] Preferably, the sub-aperture division of the spatiotemporally aligned radar data block includes: Based on the antenna configuration parameters of the MIMO array, the effective synthetic aperture length is calculated. According to the resolution requirements, the synthetic aperture length is divided into multiple overlapping sub-apertures. Based on the division results, echo data corresponding to the time period of each sub-aperture is extracted from the spatiotemporally aligned radar data block to form the sub-aperture data set.

[0009] Preferably, the beamforming process performed on each of the sub-aperture data sets includes: A corresponding steering vector is constructed for each of the sub-aperture data sets. The steering vector is used to perform spatial filtering on the sub-aperture data sets. The spatially filtered data is then pulse-compressed. The pulse-compressed data is then subjected to Fourier transform along the range direction to generate the complex images with different viewpoints.

[0010] Preferably, the step of extracting the interferometric phase information of each group of complex images and performing geometric correction on the interferometric phase information using the positioning data of the BeiDou navigation system includes: From the set of complex images with different perspectives, one image is selected as the main image, and the rest are selected as secondary images. The complex correlation coefficient between the secondary images and the main image is calculated respectively. Based on the complex correlation coefficient, an initial interferometric phase map is calculated. Using the platform position and attitude data provided by the BeiDou navigation system, the geometric phase components in the initial interferometric phase map are calculated. The geometric phase component is subtracted from the initial interference phase diagram to obtain the geometrically corrected interference phase information.

[0011] Preferably, the step of unwrapping the geometrically corrected interference phase information includes: A quality map is generated from the geometrically corrected interferometric phase information to identify regions with high phase reliability. Guided by the quality map, the branching method is used to begin phase unwrapping in the regions with high phase reliability. The phase unwrapping process is gradually extended to the entire phase map, and the unwrapping result is smoothed by using the least squares principle, outputting the continuous absolute phase field.

[0012] Preferably, the millimeter-level deformation sequence of the monitored target obtained by inversion based on the absolute phase field includes: The absolute phase field is converted into deformation increment along the radar line of sight, and the deformation rate in the time series is calculated by combining the radar wavelength and the time baseline of the two observations. The deformation rate is integrated over time to generate the millimeter-level deformation sequence. The millimeter-level deformation sequence is then subjected to noise reduction filtering to highlight the true deformation signal.

[0013] Preferably, dynamically updating the imaging focusing parameters based on the temporal evolution characteristics of the millimeter-level deformation sequence includes: The temporal gradient of the millimeter-level deformation sequence is analyzed to identify the deformation acceleration period and the deformation stabilization period. During the deformation acceleration period, the imaging integration time is shortened to reduce motion blur, and during the deformation stabilization period, the imaging integration time is increased to improve the image signal-to-noise ratio. Based on the identified period characteristics, the focusing depth parameter and aperture synthesis parameter in the imaging algorithm are adjusted in real time.

[0014] Preferably, the step of performing high-resolution imaging of the subsequently input standard-format radar data block using the updated imaging focusing parameters includes: The updated imaging focusing parameters are loaded into the imaging processing engine; The imaging processing engine performs range migration correction on the input standard format radar data block based on the focusing parameters, and performs precise phase compensation on the data after range migration correction. The final high-resolution target image is generated using a back projection algorithm or a frequency domain imaging algorithm. The high-resolution target image is associated with and stored with the corresponding millimeter-level deformation sequence, and the deformation monitoring results and target image are output.

[0015] Compared with the prior art, the beneficial effects of the present invention are: The spatiotemporally aligned radar data blocks are divided into sub-aperture datasets. The sub-aperture division is optimized based on radar beam characteristics and resolution requirements, balancing aperture length and the number of viewing angles. Each sub-aperture dataset contains echo signals within a specific viewing angle range, and overlapping areas are preserved between datasets to ensure continuity. The division process employs an adaptive algorithm, adjusting parameters based on signal-to-noise ratio and scene complexity to provide a data foundation for multi-view interferometry.

[0016] Beamforming is applied to each sub-aperture dataset to obtain a set of multi-view complex images. Digital beamforming algorithms are used for beamforming to precisely control the antenna pattern, and the processing considers array geometry and signal propagation characteristics. The complex images retain the target's amplitude and phase information, providing multi-angle observation data. Image registration algorithms eliminate geometric distortions and ensure spatial consistency. Phase information is extracted from the multi-view images through interferometry, and geometric errors are corrected using BeiDou positioning data to improve deformation monitoring accuracy.

[0017] The corrected interferometric phase is unwrapped to obtain the absolute phase field. Phase unwrapping employs a network flow or minimum norm algorithm, incorporating a quality map to guide the unwrapping process and prevent error propagation. The absolute phase field establishes a direct mathematical relationship with the deformation, providing input for millimeter-level deformation inversion. Based on the deformation sequence inversion from the absolute phase field, atmospheric and orbital errors are separated using the principle of time-series interferometry, achieving millimeter-level accuracy. The deformation sequence reveals the target's variation over time, providing data support for structural health monitoring.

[0018] Imaging focusing parameters are dynamically updated based on the temporal evolution characteristics of the deformation sequence. Temporal characteristics include deformation rate, trend, and periodic changes, and parameter updates are designed based on the mapping relationship between deformation and imaging quality. Focusing parameters include key indicators such as resolution, integration time, and beam pointing, and the adjustment process considers the system's real-time requirements. Dynamic updates ensure continuous optimization of imaging performance, forming a closed-loop feedback loop of "deformation monitoring - parameter update - imaging optimization." The updated parameters are used to perform high-resolution imaging of subsequent data blocks, outputting clear deformation monitoring results and target images, thus improving the reliability and adaptability of the monitoring system. Attached Figure Description

[0019] Figure 1 This is a schematic diagram illustrating the working principle of the high-precision deformation monitoring and target imaging method based on MIMO-SAR described in this invention. Figure 2 A flowchart for data standardization preprocessing; Figure 3 This is a flowchart of the interferometric phase extraction and geometric correction process; Figure 4 This is an analysis diagram of the phase unwrapping results; Figure 5 This is a graph showing the time evolution of the deformation sequence. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1This invention provides a high-precision deformation monitoring and target imaging method based on MIMO-SAR. The method includes: receiving raw MIMO-SAR echo data streams from an image sensor and positioning data streams from a BeiDou navigation system; performing data standardization preprocessing on the raw MIMO-SAR echo data streams to generate standard-format radar data blocks; spatiotemporally aligning the standard-format radar data blocks with the positioning data streams from the BeiDou navigation system to establish a mapping relationship between each radar data block and its corresponding spatial location; dividing the spatiotemporally aligned radar data blocks into sub-apertures to generate multiple sub-aperture data sets; and processing each sub-aperture data block separately. The sub-aperture data set is processed by beamforming to obtain a set of complex images with different perspectives. The interferometric phase information of each set of complex images is extracted, and the interferometric phase information is geometrically corrected using positioning data from the BeiDou Navigation Satellite System. The geometrically corrected interferometric phase information is unwrapped to obtain the absolute phase field. Based on the absolute phase field, the millimeter-level deformation sequence of the monitored target is obtained by inversion. The imaging focusing parameters are dynamically updated according to the temporal evolution characteristics of the millimeter-level deformation sequence. The updated imaging focusing parameters are used to perform high-resolution imaging on subsequent standard-format radar data blocks, and the deformation monitoring results and target images are output.

[0022] Example 1: See Figure 2 In specific implementations, when performing data standardization preprocessing on the raw MIMO-SAR echo data stream, a format parsing operation is performed to separate the header information and payload data. The header information includes the acquisition time and sensor parameters, while the payload data includes the raw echo signal. In specific implementations, invalid value removal and outlier correction are performed on the payload data. Invalid value removal is achieved by identifying sampling points in the data stream with amplitudes of zero or exceeding the physical range and replacing them with the mean of adjacent valid values. Outlier correction uses a sliding window statistical method to detect points deviating from the mean by more than three standard deviations and performs linear interpolation compensation. In some embodiments, the corrected payload data undergoes amplitude normalization to ensure the data amplitude is within a preset dynamic range. The amplitude normalization process is calculated using the following formula: in: This represents the amplitude value of the original data. This represents the mean amplitude of the data block. This represents the standard deviation of the data block's magnitude. This represents the normalized data amplitude value. In practice, the amplitude-normalized data is divided into standard-format radar data blocks according to a preset time window. The preset time window is set to a fixed length based on the system sampling rate and storage requirements.

[0023] In specific implementations, when aligning standard-format radar data blocks with the positioning data stream of the BeiDou Navigation Satellite System (BDS), timestamps and three-dimensional coordinate information are extracted from the BDS positioning data stream. The timestamps are recorded in UTC format, and the three-dimensional coordinate information includes longitude, latitude, and elevation. In some embodiments, the data acquisition timestamp is extracted from the header information of the standard-format radar data block, and the data acquisition timestamp uses the same time base as the BeiDou timestamp. Optionally, a time-indexed lookup table is established, storing the BDS timestamps and three-dimensional coordinate information in chronological order. In specific implementations, the data acquisition timestamp of each radar data block is matched with the nearest BDS timestamp. The matching process is based on the minimum time difference principle, with the time difference threshold set to milliseconds. Essentially, the successfully matched BDS three-dimensional coordinate information is assigned to the corresponding radar data block, completing the spatiotemporal alignment. The spatiotemporally aligned radar data blocks are then labeled with a location tag for subsequent processing. Optionally, data blocks that fail to match are marked as invalid and excluded from subsequent steps.

[0024] Example 2: In a specific implementation, when dividing the spatiotemporally aligned radar data block into sub-apertures, the effective synthetic aperture length is calculated based on the antenna configuration parameters of the MIMO array. These antenna configuration parameters include the number of antenna elements and the array spacing. In a specific implementation, the synthetic aperture length is divided into multiple overlapping sub-apertures according to resolution requirements, which are preset by system design specifications. In some embodiments, the sub-aperture division process employs a uniform segmentation strategy, and the number of sub-apertures is determined by the integer part of the ratio of the synthetic aperture length to the minimum synthetic aperture. The specific implementation of the uniform segmentation strategy includes calculating the effective synthetic aperture length based on the antenna configuration parameters of the MIMO array. These parameters include the number of antenna elements and the array spacing. The synthetic aperture length is determined based on the radar system's geometric layout and signal propagation characteristics. The minimum synthetic aperture ratio is pre-set by system design specifications, reflecting the relationship between the minimum aperture size required to achieve basic resolution and the full aperture length. The number of sub-apertures is determined by dividing the synthetic aperture length by the minimum synthetic aperture ratio, and the integer part of this quotient is used as the segmentation criterion to ensure that the number of sub-apertures is integer and adapts to resolution requirements. During segmentation, the synthetic aperture length is uniformly divided into multiple equal-length segments, each corresponding to a time interval of one sub-aperture. A fixed overlap rate is set between sub-apertures to maintain data continuity. The segmentation results are used to extract echo data for the corresponding time period from the spatiotemporally aligned radar data blocks, forming sub-aperture data sets. Each data set stores time series and channel information in matrix form. Optionally, the sub-aperture overlap rate is set to a fixed value to achieve data continuity. In practice, echo data corresponding to each sub-aperture time period is extracted from the spatiotemporally aligned radar data blocks according to the division results, forming a sub-aperture data set. The sub-aperture data set stores time series and channel information in matrix form.

[0025] In specific implementations, when performing beamforming processing on each sub-aperture dataset, a corresponding steering vector is constructed for each sub-aperture dataset. The steering vector is calculated based on the geometric center of the sub-aperture and the beam pointing angle. In some embodiments, the steering vector is used to perform spatial filtering on the sub-aperture dataset. Spatial filtering is achieved through linear weighted combination, and the filtered output is expressed as: in: The signal vector representing the sub-aperture data set, Indicates the guide vector. This represents the conjugate transpose of the directional vector. This represents the output signal after spatial domain filtering. In practice, the spatially filtered data undergoes pulse compression, and a matched filter is used to process the echo signal. The pulse-compressed data is then subjected to a Fourier transform along the range direction to generate complex images with different viewpoints. These complex images are stored in real and imaginary part formats.

[0026] Example 3: See Figure 3 In specific implementations, when extracting the interference phase information of each set of complex images, one image is selected as the master image from a set of complex images with different viewpoints, and the rest are selected as slave images. The selection of the master image is based on the image signal-to-noise ratio and phase stability index. In specific implementations, the complex correlation coefficient between the slave images and the master image is calculated separately. The complex correlation coefficient reflects the degree of phase consistency between the image pairs. In some embodiments, an initial interference phase map is calculated based on the complex correlation coefficient. The calculation process is achieved through complex conjugate multiplication and phase extraction. Optionally, the complex correlation coefficient is calculated using the following formula: in: Represents the complex pixel values ​​of the main image. Represents the complex pixel values ​​from the image. Represents the conjugate of complex pixel values ​​in an image. Indicates the total number of pixels in the image. This represents the complex correlation coefficient. In practice, the geometric phase component in the initial interferometric phase map is calculated using platform position and attitude data provided by the BeiDou Navigation Satellite System. This geometric phase component is derived from the path difference caused by platform motion. It can be understood that subtracting the geometric phase component from the initial interferometric phase map yields geometrically corrected interferometric phase information, which focuses on the phase change caused by target deformation.

[0027] In specific implementations, when unwrapping the geometrically corrected interferometric phase information, a quality map is generated. This quality map identifies regions with high phase reliability based on phase gradient and coherence indices. In some embodiments, guided by the quality map, a branch-cutting method is used to begin phase unwrapping in regions with high phase reliability. This method avoids path integral errors by setting branch tangents. In specific implementations, the phase unwrapping process is progressively extended to the entire phase map, following the reliability order indicated by the quality map. Optionally, the unwrapping result is smoothed using the least squares principle, achieved by minimizing phase gradient differences. It can be understood that phase unwrapping outputs a continuous absolute phase field, which is used for subsequent deformation inversion processing. The setting of branch tangents during phase unwrapping is based on phase residual point detection.

[0028] See Figure 4This paper presents the absolute phase field distribution results obtained through interferometric synthetic aperture radar (IAPR) technology. The figure shows the continuous phase distribution after geometric correction and phase unwrapping, obtained using a branch-cut phase unwrapping algorithm based on complex image interferometry. The color distribution in the graph represents the phase values ​​of different regions, with the gradient from blue to red reflecting the continuous change in phase values. This continuous phase field is the foundation for subsequent deformation monitoring; the phase value of each pixel is directly related to the path change along the radar line of sight. A distinct phase fringe pattern can be observed in the figure, reflecting minute surface deformation information. The continuity and consistency of the phase values ​​verify the effectiveness of the phase unwrapping algorithm, providing reliable input data for subsequent deformation inversion. This phase field distribution map demonstrates the effective recovery of the interferometric phase by the processing algorithm, eliminating phase entanglement and obtaining true absolute phase values. This processing is crucial for achieving millimeter-level precision deformation monitoring, providing technical support for infrastructure safety monitoring and geological disaster early warning.

[0029] Example 4: In a specific implementation, when obtaining the millimeter-level deformation sequence of the monitored target based on the absolute phase field inversion, the absolute phase field is converted into deformation increments along the radar line-of-sight direction. The conversion process is achieved through the linear relationship between phase and deformation, and the radar line-of-sight direction is determined by the sensor's geometric configuration. In some embodiments, the deformation rate on the time series is calculated by combining the radar wavelength and the time baseline of two observations. The deformation rate is calculated using the following formula: in: Indicates the deformation rate. Indicates the radar wavelength. This represents the change in the absolute phase field. This represents the time baseline between two observations. In practice, the deformation rate is integrated over time to generate a millimeter-level deformation sequence. The time-series integration uses an accumulation method to process the deformation rate values ​​at consecutive time points. It can be understood that denoising filtering is applied to the millimeter-level deformation sequence to highlight the true deformation signal; wavelet thresholding is used for denoising. Optionally, the deformation sequence is stored in a format corresponding to timestamps and deformation amounts.

[0030] In specific implementations, when dynamically updating imaging focusing parameters based on the temporal evolution characteristics of millimeter-level deformation sequences, the temporal gradient of the millimeter-level deformation sequence is analyzed. The temporal gradient is obtained by calculating the difference in deformation at adjacent time points. In some embodiments, deformation acceleration periods and deformation stabilization periods are identified. Deformation acceleration periods correspond to intervals where the temporal gradient exceeds a preset threshold, while deformation stabilization periods correspond to intervals where the temporal gradient is below a preset threshold. In specific implementations, the imaging integration time is shortened during deformation acceleration periods to reduce motion blur, and the adjustment of the imaging integration time is based on the magnitude of the deformation rate change. Optionally, the imaging integration time is increased during deformation stabilization periods to improve the image signal-to-noise ratio, with the increase in integration time proportional to the degree of deformation stabilization. It can be understood that, referring to Table 1, the focusing depth parameter and aperture synthesis parameter in the imaging algorithm are adjusted in real time according to the identified time period characteristics. The focusing depth parameter is calculated based on the platform distance and resolution requirements, and the aperture synthesis parameter is updated based on the sub-aperture division results.

[0031] Table 1: Parameter Mapping Table In practice, the parameter mapping table is predefined based on the system calibration data, and the reference value is determined by the radar system performance indicators.

[0032] See Figure 5 This figure demonstrates the temporal evolution characteristics of millimeter-scale deformation sequences obtained based on absolute phase field inversion. The three curves in the figure represent the deformation processes of regions with different deformation modes during the monitoring period, covering various deformation behaviors such as stable subsidence, accelerated deformation, and periodic deformation. The blue curve shows a relatively stable subsidence process, with the deformation rate remaining essentially constant, reflecting the continuous ground subsidence in the region. The red curve shows obvious accelerated deformation characteristics, with a significant increase in the deformation rate in the later stages of monitoring, potentially indicating potential geological risks. The green curve exhibits a typical periodic deformation pattern, closely related to seasonal factors or changes in external loads. Through temporal gradient analysis of these deformation sequences, the system can automatically identify periods of accelerated and stable deformation. This identification provides a basis for the dynamic adjustment of imaging parameters: shortening the imaging integration time during periods of accelerated deformation to reduce motion blur, and increasing the integration time during stable periods to improve the image signal-to-noise ratio. Accurate monitoring of deformation sequences provides crucial data support for infrastructure safety assessment, geological disaster early warning, and urban subsidence monitoring, enabling refined management and control of surface deformation processes.

[0033] Example 5: In a specific implementation, the updated imaging focusing parameters are loaded into the imaging processing engine. The imaging processing engine is a data processing unit integrating hardware and software. The updated imaging focusing parameters include focusing depth parameters and aperture synthesis parameters, which are acquired from the dynamic update process and stored in the parameter cache. In a specific implementation, the imaging processing engine performs range migration correction on the input standard format radar data blocks based on the focusing parameters. Range migration correction is achieved by compensating for the range offset of the echo signal caused by platform motion. The correction process uses an interpolation algorithm to adjust the coordinates of the range cells in the radar data blocks. In some embodiments, precise phase compensation is performed on the data after range migration correction. The phase compensation is based on the phase error model in the focusing parameters, which is derived from the platform motion history and radar system calibration data. During the derivation process, by analyzing the position fluctuations and attitude drift phenomena in the platform motion history and combining the parameter deviations in the radar system calibration data, the influence of unsatisfactory platform motion and system hardware errors on the echo signal phase is evaluated, thereby establishing a correlation model between phase error and motion state and system parameters. This model is used to accurately estimate the phase error value after range migration correction. Optionally, the phase compensation is calculated using the following formula: in: This indicates the compensated phase. This represents the original phase after distance migration correction. This represents the phase error estimate obtained from the focusing parameters. In specific implementations, the final high-resolution target image is generated using a back projection algorithm or a frequency domain imaging algorithm. The back projection algorithm is achieved by back-projecting the echo data of each aperture point onto an image grid and coherently superimposing them. The frequency domain imaging algorithm uses a range-Doppler algorithm or a wavenumber domain algorithm to process the phase-compensated data. It can be understood that the high-resolution target image is stored in the form of a complex matrix, containing amplitude and phase components. In some embodiments, the high-resolution target image is associated with the corresponding millimeter-level deformation sequence. This association is accomplished by matching timestamps and spatial coordinates, outputting the deformation monitoring results and the target image. The deformation monitoring results include metadata for the deformation sequence data and the high-resolution target image. Optionally, the association uses a database indexing mechanism to ensure data consistency. In specific implementations, the output process packages the deformation monitoring results and the target image into a standardized data format for subsequent analysis or visualization.

[0034] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0035] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-precision deformation monitoring and target imaging method based on MIMO-SAR, characterized in that, Includes the following steps: Receive raw MIMO-SAR echo data stream from the image sensor and positioning data stream from the BeiDou Navigation Satellite System; The original MIMO-SAR echo data stream is preprocessed using data standardization to generate radar data blocks in a standard format; The standard format radar data blocks are spatiotemporally aligned with the positioning data stream of the BeiDou navigation system to establish a mapping relationship between each radar data block and its corresponding spatial location. The spatiotemporally aligned radar data blocks are divided into sub-apertures to generate multiple sub-aperture datasets. Beamforming processing is performed on each of the sub-aperture data sets to obtain a set of complex images with different viewpoints; The interference phase information of each group of complex images is extracted, and the interference phase information is geometrically corrected using the positioning data of the Beidou navigation system; The geometrically corrected interference phase information is unwrapped to obtain the absolute phase field; Based on the absolute phase field, the millimeter-level deformation sequence of the monitored target is obtained by inversion; Based on the temporal evolution characteristics of the millimeter-level deformation sequence, the imaging focusing parameters are dynamically updated; Using the updated imaging focusing parameters, high-resolution imaging is performed on the subsequently input standard format radar data blocks, and deformation monitoring results and target images are output.

2. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 1, characterized in that, The data normalization preprocessing of the raw MIMO-SAR echo data stream includes: The original MIMO-SAR echo data stream is parsed to separate the header information and payload data; Invalid values ​​are removed and outliers are corrected in the payload data; The amplitude of the corrected payload data is normalized so that the data amplitude is within the preset dynamic range; The amplitude-normalized data is divided into segments according to a preset time window to generate radar data blocks in the standard format.

3. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 2, characterized in that, The step of aligning the standard format radar data block with the positioning data stream of the BeiDou navigation system in time and space includes: Extract timestamps and three-dimensional coordinate information from the positioning data stream of the BeiDou navigation system; Extract the data acquisition timestamp from the header information of the radar data block in the standard format; A time-indexed lookup table is established to match the data acquisition timestamp of each radar data block with the closest timestamp of the BeiDou navigation system. The three-dimensional coordinate information of the successfully matched BeiDou navigation system is assigned to the corresponding radar data block to complete the spatiotemporal alignment.

4. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 1, characterized in that, The sub-aperture division of the spatiotemporally aligned radar data block includes: Based on the antenna configuration parameters of the MIMO array, the effective synthetic aperture length is calculated. According to the resolution requirements, the synthetic aperture length is divided into multiple overlapping sub-apertures. Based on the division results, echo data corresponding to the time period of each sub-aperture is extracted from the spatiotemporally aligned radar data block to form the sub-aperture data set.

5. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 4, characterized in that, Beamforming processing for each of the sub-aperture data sets includes: A corresponding steering vector is constructed for each of the sub-aperture data sets. The steering vector is used to perform spatial filtering on the sub-aperture data sets. The spatially filtered data is then pulse-compressed. The pulse-compressed data is then subjected to Fourier transform along the range direction to generate the complex images with different viewpoints.

6. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 1, characterized in that, Extracting the interferometric phase information of each group of complex images and performing geometric correction on the interferometric phase information using the positioning data of the BeiDou navigation system includes: From the set of complex images with different perspectives, one image is selected as the main image, and the rest are selected as secondary images. The complex correlation coefficient between the secondary images and the main image is calculated respectively. Based on the complex correlation coefficient, an initial interferometric phase map is calculated. Using the platform position and attitude data provided by the BeiDou navigation system, the geometric phase components in the initial interferometric phase map are calculated. The geometric phase component is subtracted from the initial interference phase diagram to obtain the geometrically corrected interference phase information.

7. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 6, characterized in that, The step of unwrapping the geometrically corrected interferometric phase information includes: A quality map is generated from the geometrically corrected interferometric phase information to identify regions with high phase reliability. Guided by the quality map, the branching method is used to begin phase unwrapping in the regions with high phase reliability. The phase unwrapping process is gradually extended to the entire phase map, and the unwrapping result is smoothed by using the least squares principle, outputting the continuous absolute phase field.

8. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 1, characterized in that, Based on the absolute phase field, the millimeter-level deformation sequence of the monitored target is inverted, including: The absolute phase field is converted into deformation increment along the radar line of sight, and the deformation rate in the time series is calculated by combining the radar wavelength and the time baseline of the two observations. The deformation rate is integrated over time to generate the millimeter-level deformation sequence. The millimeter-level deformation sequence is then subjected to noise reduction filtering to highlight the true deformation signal.

9. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 8, characterized in that, The step of dynamically updating the imaging focusing parameters based on the time evolution characteristics of the millimeter-level deformation sequence includes: The temporal gradient of the millimeter-level deformation sequence is analyzed to identify the deformation acceleration period and the deformation stabilization period. During the deformation acceleration period, the imaging integration time is shortened to reduce motion blur, and during the deformation stabilization period, the imaging integration time is increased to improve the image signal-to-noise ratio. Based on the identified period characteristics, the focusing depth parameter and aperture synthesis parameter in the imaging algorithm are adjusted in real time.

10. The high-precision deformation monitoring and target imaging method based on MIMO-SAR according to claim 1, characterized in that, The step of performing high-resolution imaging of the subsequently input standard-format radar data blocks using the updated imaging focusing parameters includes: The updated imaging focusing parameters are loaded into the imaging processing engine; The imaging processing engine performs range migration correction on the input standard format radar data block based on the focusing parameters, and performs precise phase compensation on the data after range migration correction. The final high-resolution target image is generated using a back projection algorithm or a frequency domain imaging algorithm. The high-resolution target image is associated with and stored with the corresponding millimeter-level deformation sequence, and the deformation monitoring results and target image are output.

Images