A GB-ArcSAR tomographic imaging method

Abstract

The application discloses a GB-ArcSAR tomographic imaging method. Radar images with opposite pitch angles are selected as an interferogram image pair, a vertical and uniform interference baseline is obtained, and the relationship between the interference phase and the target height is simplified, then combined with tomographic measurement point selection and height spectrum estimation, under GB-ArcSAR arc-shaped height synthetic aperture, the acquisition of the tomographic point cloud is realized.

Description

Technical Field

[0001] This invention relates to the field of synthetic aperture radar technology, and more specifically to a GB-ArcSAR tomographic imaging method. Background Technology

[0002] Synthetic Aperture Radar (SAR) has been widely used in vegetation vertical structure inversion, surface monitoring, glacier movement monitoring, and urban dynamic deformation monitoring. Compared with airborne and spaceborne SAR, Ground-Based Synthetic Aperture Radar (GB-SAR) has real-time monitoring capabilities and can flexibly adjust the observation angle.

[0003] SAR tomography utilizes a platform equipped with multiple antennas for single-pass or single-antenna-multiple-pass observations of the same area, forming a synthetic aperture in the altitude direction. This overcomes the limitations of the superposition effect, enabling effective separation of multiple scatterers within a single resolution cell, thus giving SAR range, azimuth, and altitude resolution capabilities. GB-ArcSAR (Ground-Based Arc Synthetic Aperture Radar) acquires an arc-shaped synthetic aperture by controlling the transceiver antennas to scan in a horizontal plane using a turntable, providing 360° full-field-of-view imaging capability in the azimuth direction.

[0004] To enable GB-ArcSAR to perform tomographic imaging, a two-dimensional turntable is used to control the transceiver antenna to scan in multiple horizontal planes at different altitudes. Unlike spaceborne or airborne tomographic SAR, the synthetic apertures of GB-ArcSAR in both the azimuth and elevation directions are curved, and there is currently no tomographic imaging method directly applicable to GB-ArcSAR.

[0005] Therefore, research on GB-ArcSAR tomographic imaging methods is needed. Summary of the Invention

[0006] In view of this, the present invention provides a GB-ArcSAR tomographic imaging method, which achieves the acquisition of tomographic point clouds through interferometric baseline design, tomographic measurement point selection, and height spectral estimation.

[0007] The GB-ArcSAR tomographic imaging method of the present invention includes:

[0008] Step 1: Determine the number of altitude sampling points N and the GB-ArcSAR elevation angle θ. i , i = 1, 2, ..., N; where the number of altitude sampling points N is determined by the radar resolution and the maximum unambiguous altitude; under a uniform baseline, the altitude pitch angle θi (i∈1,2,...,N) is determined based on the baseline interval Δh and the number of height sampling points N;

[0009] Step 2, adjust the GB-ArcSAR elevation angle to -θ N ,-θ N-1 ,...,-θ1,θ1,...,θ N-1 ,θ N By scanning the same observation scene in the horizontal plane, 2N radar images are obtained;

[0010] Step 3: Filter the pixels in the 2N radar images to obtain the tomographic measurement points;

[0011] Step 4, select θ i and -θ i Interference is performed on two images obtained at a certain location to obtain N interferograms;

[0012] Step 5: Extract the complex interferometric sequence of the tomographic measurement point in N interferograms, perform height spectral estimation to obtain the height estimate of the tomographic measurement point, and then combine it with the range-azimuth information of the tomographic measurement point in the radar image to obtain the three-dimensional tomographic point cloud and complete the tomographic imaging.

[0013] Preferably, in step 1, the relationship between radar resolution ρ and baseline length L is as follows:

[0014]

[0015] Where λ is the radar signal wavelength, R is the distance between the target and the radar; and the maximum unambiguous height h u The relationship with the baseline interval Δh is as follows:

[0016]

[0017] The number of height sampling points N satisfies:

[0018]

[0019] Preferably, in step 1, θ i The possible values ​​are as follows:

[0020]

[0021] Where Δh is the height baseline interval; r is the GB Arc-SAR rotation radius.

[0022] Preferably, in step 3, a correlation coefficient threshold γ is used. T Screening of chromatographic measurement points.

[0023] Preferably, in step 3, for each pixel, the coherence coefficient of that pixel in adjacent image pairs is calculated, and then the mean of the coherence coefficients is obtained; if the mean is greater than the set correlation coefficient threshold γ... T If so, then the pixel is the selected tomographic measurement point.

[0024] Preferably, in step 5, the interferometric complex sequence is subjected to IDFT processing to achieve height spectral estimation.

[0025] Beneficial effects:

[0026] This invention selects radar images with opposite pitch angles as interferometric image pairs to obtain a vertical and uniform interferometric baseline, simplifying the relationship between the interferometric phase and the target height. Then, by combining the selection of tomographic measurement points and the height spectrum estimation, tomographic point cloud acquisition is achieved under the GB-ArcSAR arc height synthetic aperture. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the GB-ArcSAR tomography measurement method.

[0028] Figure 2 This is a schematic diagram of GB-ArcSAR interferometry on the range-elevation plane.

[0029] Figure 3 This is a flowchart of the method of the present invention. Detailed Implementation

[0030] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0031] This invention provides a GB-ArcSAR tomographic imaging method, the flowchart of which is shown below. Figure 3 As shown, the specific steps include the following:

[0032] Step 1: Determine the number of height sampling points N and the height rotation angle θ i

[0033] To enable GB-ArcSAR to perform tomographic imaging, a two-dimensional turntable is needed to control a single pair of transceiver antennas to scan at multiple horizontal planes at different heights. Figure 1 This is a schematic diagram of the scanning method for GB-ArcSAR tomography. For example... Figure 1 As shown, the radar performs circular scanning on the horizontal plane at different altitudes, forming an arc-shaped synthetic aperture in the azimuth and altitude directions.

[0034] To simplify the chromatographic process, this invention employs a uniform baseline. Let the required resolution be ρ, and the maximum unambiguous height be h. uThe radar uses a wavelength of λ, the slant range between the target and the radar is R, the number of altitude sampling points is N, and the altitude baseline spacing is Δh. The relationship between the resolution ρ and the baseline length L is:

[0035]

[0036] Maximum unambiguous height h u The relationship with the baseline interval Δh can be expressed as:

[0037]

[0038] Under the conditions of satisfying the required resolution and the maximum unblurred height, the number of height sampling points N satisfies:

[0039]

[0040] Figure 2 The diagram shows an interferometric measurement on the yz plane. Gray dots represent the scan center, blue dots represent the height sampling positions, and the blue dashed line represents the cantilever length, i.e., the GB Arc-SAR rotation radius r. Assume the two sampling positions are S1 and S2, with elevation angles of ω1 and ω2, and coordinates of (rcosω1, rsinω1) and (rcosω2, rsinω2), respectively. The coordinates of the point target P are (y... P ,z P The pitch angle is θ. P The slant range from the radar is R. P The slope distance difference ΔR between S1, S2 and the target point P during the two measurements can be expressed as:

[0041]

[0042] When ω1 ≠ -ω2, the interference baseline will include both horizontal and vertical components. The presence of the horizontal component increases the difficulty of tomographic processing. To simplify the process, we can let ω2 = -ω1, in which case the interference baseline will be vertical.

[0043] Point target P interferometric phase The relationship between it and its height h can be simplified as follows:

[0044]

[0045] Let θ i (i = 1, 2, ..., N) represents the pitch angle from altitude. Δh = 2rsinθ1, taking θ1 = ω1, we can calculate... To satisfy the uniform baseline condition, the change in radar antenna height must remain constant after each elevation angle adjustment, meaning the calculated Δh must remain unchanged. i The possible values ​​are as follows:

[0046]

[0047] Step 2: Radar Image Acquisition

[0048] As can be seen from step one, when the pitch angles are opposite numbers, the interference baseline is vertical.

[0049] Adjust the pitch angle to -θ N ,-θ N-1 ,...,-θ1,θ1,...,θ N-1 ,θ N By scanning the same observation scene within the horizontal plane, 2N echo data files are obtained. The echo data can be processed using a back projection algorithm to obtain a two-dimensional radar imaging result, i.e., an imaging result containing range-azimuth information.

[0050] Step 3: Screening of Chromatographic Measurement Points

[0051] Using the correlation coefficient threshold γ T Selecting tomographic measurement points. Coherence is one of the intuitive standards for measuring interferometric phase noise. Given a window with a certain radius (e.g., radius 3, 5, etc.), calculate the coherence coefficient of this pixel within a certain spatial range as follows:

[0052]

[0053] Where M and S are the main image and auxiliary image, respectively, p and q are the number of rows and columns of pixels in the image, and m and n represent the number of rows and columns of the rectangular window. The coherence coefficient of pixels can be calculated using adjacent images: select the first image as the main image and the second image as the auxiliary image to calculate γ1; then select the second image as the main image and the third image as the auxiliary image to calculate γ2, and so on. Using the 2N radar images obtained in step two, the coherence coefficients (γ1, γ2, ..., γ) can be calculated. 2N-2 ,γ 2N-1 To select the measurement points for chromatographic analysis, the average coherence coefficient is calculated pixel by pixel:

[0054]

[0055] like Then select it as the chromatographic measurement point. γ T The typical value range is 0.7 to 0.9.

[0056] Step 4: Obtaining the interferogram

[0057] Let θ i The radar image matrix obtained at point P is i , at -θ i The radar image matrix obtained at point P isi- , for P i With P i- Perform complex conjugate multiplication to obtain interferogram I. i :

[0058]

[0059] Where * denotes the conjugate of a complex number.

[0060] Therefore, for 2N radar images, N interferometric phase maps can be obtained.

[0061] Taking the point target P in step one as an example, its interference phase sequence in N interferometric phase diagrams can be represented as:

[0062]

[0063] Step 5: Height-oriented spectrum estimation

[0064] For each tomographic measurement point in step three, taking point target P as an example, the complex interferometric sequence of it in N interferometric phase maps is extracted as follows: By performing discrete inverse Fourier transform (IDFT) on the complex sequence to estimate the height spectrum, we can obtain:

[0065]

[0066] Suppose S(k) is at k max If the maximum value is obtained at a certain point, then the estimated height of that chromatographic measurement point is [value missing]. By combining the range-azimuth information of each tomographic measurement point in the radar image, a three-dimensional tomographic point cloud can be obtained.

[0067] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A GB-ArcSAR tomographic imaging method, characterized in that, include: Step 1: Determine the number of altitude sampling points N and the GB-ArcSAR elevation angle. , i =1,2,…, N The number of altitude sampling points N is determined by the radar resolution and the maximum unambiguous altitude; under a uniform baseline, the altitude pitch angle... According to baseline interval The number of height-based sampling points N is calculated and determined. Among them, radar resolution With baseline length L The relationship is: in, R is the radar signal wavelength, and R is the distance between the target and the radar; maximum unambiguous altitude. Baseline Interval The relationship is: The number of height sampling points N satisfies: The possible values ​​are as follows: in, r is the height-to-baseline interval; r is the GB Arc-SAR rotation radius; Step 2, adjust the GB-ArcSAR elevation angle as follows: By scanning the same observation scene in the horizontal plane, 2N radar images are obtained; Step 3: Filter the pixels in the 2N radar images to obtain tomographic measurement points; wherein, for each pixel, calculate the coherence coefficient of that pixel in adjacent image pairs, and then calculate the mean of the coherence coefficients; if the mean is greater than a set correlation coefficient threshold... If so, then the pixel is the selected tomographic measurement point; Step 4, select and Interference is performed on two images obtained at a certain location to obtain N interferograms; Step 5: Extract the complex interferometric sequence of the tomographic measurement point in N interferograms, perform height spectral estimation to obtain the height estimate of the tomographic measurement point, and then combine it with the range-azimuth information of the tomographic measurement point in the radar image to obtain the three-dimensional tomographic point cloud and complete the tomographic imaging.

2. The method as described in claim 1, characterized in that, In step 5, the interferometric complex sequence is processed by IDFT to achieve height spectral estimation.

Images