A new method for estimating scattering characteristics of synthetic aperture radar corner reflectors

Abstract

This invention belongs to the field of space-based Earth observation and proposes a new method for estimating the scattering characteristics of synthetic aperture radar corner reflectors. This method utilizes SAR imagery before CR installation, eliminating interference from CR reflection signals in estimating true background intensity. It also selects pixels with the same statistical distribution within a sampling window and calculates their average value as the background intensity, effectively eliminating anomalous pixels within the window. Furthermore, it fully utilizes the spatial distribution characteristics of CR reflection signals in SAR images, facilitating the extraction of the CR's true reflection intensity. This method is applicable to various scenarios and can obtain RCS and SCR values that are more realistic.

Description

Technical Field

[0001] The present invention relates to the field of space earth observation, and in particular to a new method for estimating scattering characteristics of a synthetic aperture radar corner reflector. Background Art

[0002] Synthetic Aperture Radar (SAR) is a new space-based Earth observation technology with unique advantages, including all-weather, global coverage, high temporal and spatial resolution, high precision, and automated processing. Ground-mounted corner reflectors (CRs) are ideal impulse response targets, appearing as bright spots in SAR images and easily identifiable. They are widely used in SAR image calibration and surface deformation monitoring.

[0003] The backscatter characteristics of a CR can be evaluated using two metrics: radar cross section (RCS) and signal-to-clutter ratio (SCR). The RCS reflects the intensity of a CR's reflection in SAR imagery, while the SCR reflects its brightness. Its value depends on both the amount of its own reflected energy and the amount of background reflected energy. The SCR value can, to a certain extent, predict the stability of the interferometric phase variation of a point target and the absolute positioning accuracy. A higher SCR value increases the likelihood of achieving high-precision positioning results. Commonly used methods for estimating RCS and SCR values can be divided into two categories: one based on spatial information, calculating the CR signal from neighboring pixels, and the other based on the statistics of the CR reflection intensity time series. The first method is more suitable for high-resolution imagery and large CRs, where the CR signal appears as a distinct cross in SAR imagery, making it easier to distinguish CR reflected energy from background reflected energy. The second method is more suitable for complex environments, the presence of other strong point scatterers, small CRs, and medium- and low-resolution SAR imagery. For example, the low spatial resolution of medium-resolution Sentinel-1IW SAR images makes it easier for neighboring pixels to contain strong reflection interference from point targets other than corner reflectors. In this case, the second method can obtain an estimate that is closer to the actual situation.

[0004] Combining the two different algorithms, we can see that the key to unbiased estimation of RCS and SCR values lies in how to obtain background and CR reflection intensities that are more consistent with the actual situation. The spatial method is based on the assumption that the scattering environment within the region is similar. If there is contamination from other strong reflective targets, the true environmental reflection intensity cannot be estimated. The temporal method is based on the premise that the background reflection intensity is stable in the time series. However, this method only uses a single pixel (intensity peak) as the statistic. Due to the limited bandwidth of the radar acquisition signal, sidelobes are generated when the SINC function is used to convert it into a time domain signal. Therefore, the CR reflection signal in SAR images is usually distributed among several adjacent pixels. Using only a single pixel cannot reflect the true CR reflection intensity, and the calculated RCS value is too small. Summary of the Invention

[0005] The present invention is made to solve the above problems and aims to provide a new method for estimating the scattering characteristics of a synthetic aperture radar corner reflector. The present invention adopts the following technical solutions:

[0006] The present invention provides a new method for estimating the scattering characteristics of a synthetic aperture radar corner reflector, which is characterized by comprising the following steps: step S1, using multiple SAR images without a CR installed, and using the geographic coordinates of the CR to inversely calculate the pixel position of the CR in the image; step S2, taking the inversely calculated pixel position as the center, intercepting a sub-block of a window of appropriate size in the intensity image without the CR installed, and performing absolute correction, thereby obtaining the backscattering intensity value σ of each pixel 0 Step S3, the backscattering intensity value σ of the central pixel of the sub-block window 0 The time series of the pixels in the sub-block window is used as the reference sample, and the time series samples of the remaining pixels in the sub-block window are tested using the two-sample KS hypothesis test method; step S4, the pixel that passes the test is regarded as a SHP, and the average intensity value E of all the SHP points is calculated. SHP , thereby estimating the background reflection intensity; Step S5, using the SAR image after installing CR, taking the pixel where the CR intensity peak is located as the center, selecting a target window of a predetermined size, and deducting the background intensity of the target window, thereby obtaining the true reflection intensity E of the CR CR ; Step S6, based on the average intensity value E SHP and the reflection intensity E CR The RCS value and the SCR value are calculated, and the temporal stability of the RCS value is analyzed to evaluate the stability of the CR state and exclude abnormal data.

[0007] The new method for estimating the scattering characteristics of a synthetic aperture radar corner reflector provided by the present invention may also have such a technical feature, wherein step S3 is to determine whether the time series sample and the reference sample come from the same distribution, thereby determining whether the time series sample pixel passes the inspection.

[0008] The new method for estimating the scattering characteristics of synthetic aperture radar corner reflectors provided by the present invention may also have the following technical features: wherein the true reflection intensity E of the CR in step S5 is CR The calculation formula is as follows:

[0009]

[0010] Where N represents the number of pixels contained in the window, is the backscattering intensity value of the window pixel, E SHP is the average intensity value.

[0011] The new method for estimating the scattering characteristics of a synthetic aperture radar corner reflector provided by the present invention may also have the following technical features: wherein the calculation formulas for the RCS value and the SCR value in step S6 are as follows:

[0012]

[0013]

[0014] Where, ρ rg represents the orientation of the image, ρ az Represents the distance to pixel interval, θ loc It is expressed as a local incident angle, and the abnormal data includes water accumulation and damage of the CR.

[0015] Functions and effects of the invention

[0016] The new method for estimating the scattering characteristics of synthetic aperture radar corner reflectors (SAR) uses SAR images taken before CR installation, eliminating interference from CR reflection signals in estimating true background intensity. Furthermore, the method selects pixels with the same statistical distribution within a sampling window and calculates their average value as the background intensity, effectively eliminating anomalous pixels within the window. Furthermore, CR reflection energy is typically distributed across several surrounding pixels. This method uses this spatial distribution to select an appropriate window for estimating CR reflection intensity, resulting in a more realistic result. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 1 is a flow chart of a new method for estimating scattering characteristics of a synthetic aperture radar corner reflector according to an embodiment of the present invention;

[0018] Figure 2Schematic diagram of the RCS of a corner reflector changing with time in an embodiment of the present invention;

[0019] Figure 3 Schematic diagram of the calculation results of RCS and SCR values of the corner reflector in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the technical means, creative features, objectives and effects of the present invention easier to understand, the new method for estimating the scattering characteristics of a synthetic aperture radar corner reflector of the present invention is specifically described below with reference to embodiments and drawings.

[0021] <Example>

[0022] Figure 1 4 is a flow chart of a new method for estimating scattering characteristics of a synthetic aperture radar corner reflector in an embodiment of the present invention.

[0023] like Figure 1 As shown in Figure 2, the new method for estimating the scattering characteristics of synthetic aperture radar corner reflectors mainly includes the following steps:

[0024] Step S1: Using a SAR image without a CR installed, and using the geographic coordinates of the CR to inversely calculate its pixel position in the image.

[0025] In this embodiment, SAR images without CR (the aforementioned radar corner reflector) are used. The number of SAR images without CR is preferably greater than 20, and the pixel positions in the images are inversely calculated using the CR geographic coordinates. SAR images include intensity images.

[0026] Step S2: With the inversely calculated pixel position as the center, a sub-block of a suitable size window is intercepted from the SAR image (intensity image) without CR installed, and absolute correction is performed to obtain the backscatter intensity value σ of each pixel. 0 .

[0027] Step S3: taking the backscatter intensity value σ of the central pixel of the sub-block window as the value 0 The time series of is taken as the reference sample, and the time series samples of the remaining pixels in the sub-block window are tested using the two-sample KS hypothesis test method.

[0028] In this embodiment, the σ of the central pixel of the sub-block is 0 The time series of the value is used as the reference sample, and the two-sample KS hypothesis test method is used to determine whether the time series samples of the remaining pixels in the window and the reference sample come from the same distribution. The pixels that pass the hypothesis test can be identified as a SHP;

[0029] Step S4: consider the pixel that has passed the test as a SHP, and calculate the average intensity value E of all the SHP points.SHP , thereby estimating the background reflection intensity;

[0030] In this embodiment, the average intensity value E of all SHP points is calculated. SHP It is approximated to the background signal intensity, and the number of the SHP points is counted, which can be used as a reliability indicator of the RCS value and SCR value calculated subsequently. The more the number of SHP points, the closer the calculated RCS value and SCR value are to the actual situation.

[0031] Step S5: Using the SAR image after installing the CR, taking the pixel where the CR intensity peak is located as the center, selecting a target window of a predetermined size, and deducting the background intensity of the target window, thereby obtaining the true reflection intensity E of the CR. CR .

[0032] In this embodiment, the SAR image after installing CR is used, and the pixel where the CR intensity peak is located is centered. A target window of appropriate size is selected, which contains a total of N pixels. After deducting the background intensity in the window, the real reflection intensity E of CR is obtained. CR , the calculation formula is as follows:

[0033]

[0034] Where N represents the number of pixels contained in the window, is the backscattering intensity value of the window pixel, E SHP is the average intensity value.

[0035] Step S6: Based on the average intensity value E SHP and the reflection intensity E CR Calculate the RCS value and SCR value, and analyze the stability of the RCS value over time, so as to evaluate the stability of the CR state and eliminate abnormal data;

[0036] In this embodiment, the calculation formulas for the RCS value and the SCR value are as follows:

[0037]

[0038]

[0039] Where, ρ rg represents the orientation of the image, ρ az Represents the distance to pixel interval, θ loc It is expressed as a local incident angle, and the temporal stability of the RCS value is analyzed to evaluate the stability of the corner reflector state and exclude abnormal data (such as water accumulation or damage in the corner reflector).

[0040] Figure 2Schematic diagram of the RCS of a corner reflector changing with time in an embodiment of the present invention; Figure 3 Schematic diagram of the calculation results of RCS and SCR values of the corner reflector in an embodiment of the present invention.

[0041] like Figure 2-3 As shown, this embodiment uses 5 CRs with a right-angle side length of 1.2m deployed in Shanghai to test and evaluate the designed new method. The numbers and installation times of the CRs are: DT (2020.7.27), CX (2020.8.7), B1 (2020.9.7), B2 (2020.9.8) and B3 (2020.9.10). The ascending orbit Sentinel-1A / B IW mode SAR images covering 5 CRs were collected, of which 25 images without CRs installed were from September 9, 2019 to July 17, 2020; and 81 images after CRs were installed were from July 29, 2020 to March 9, 2022. According to the spatial distribution characteristics of CRs in SAR images, a 7×7 sampling window is adopted, as shown in Figure 2 The figure shows the change of RCS value of CR in each scene image with time. Figure 2 It can be seen that the RCS of the five corner reflectors increased significantly after the installation date. The larger the value, the stronger the reflection ability of the CR. Figure 2 After removing the problematic images, the average RCS and SCR values of the five groups of CR were obtained, as shown in Figure 3 As shown in the figure, the CX is installed in a parking lot, where its reflection intensity is enhanced by the occasional large vehicles parked there, and the corresponding RCS value exceeds the ideal value (34.5dB·m 2 ), and its SCR value is also too large. Excluding CX, DT, installed in the field, has the largest SCR value. B1 and B2, installed on the bridge approach, have the second-largest SCR value. B3, installed on the main bridge composite beam, has the smallest SCR value. This result is consistent with the actual situation.

[0042] Example Function and Effect

[0043] The new method for estimating the scattering characteristics of synthetic aperture radar corner reflectors (SAR) uses SAR images taken before CR installation, eliminating interference from CR reflection signals in estimating true background intensity. Furthermore, the method selects pixels with the same statistical distribution within a sampling window and calculates their average value as the background intensity, effectively eliminating anomalous pixels within the window. Furthermore, CR reflection energy is typically distributed across several surrounding pixels. This method uses this spatial distribution to select an appropriate window for estimating CR reflection intensity, resulting in a more realistic result.

[0044] In the embodiment, the designed new method was tested and evaluated using CRs with five right-angle sides of 1.2 m in Shanghai. The test results were consistent with the actual situation, proving that the new method of the present invention is effective and reliable.

[0045] The above embodiments are only used to illustrate specific implementations of the present invention, and the present invention is not limited to the description scope of the above embodiments.

Claims

1. A new method for estimating the scattering characteristics of synthetic aperture radar corner reflectors, characterized in that: The following steps are involved: Step S1, using multiple SAR images without CR installed, and using the geographic coordinates of the CR to inversely calculate its pixel position in the image; Step S2: Taking the inversely calculated pixel position as the center, a sub-block window of a suitable size is intercepted from the intensity image without CR installed, and absolute correction is performed to obtain the backscatter intensity value σ of each pixel. 0 ; Step S3: taking the backscatter intensity value σ of the central pixel of the sub-block window as the value 0 The time series of is taken as the reference sample, and the time series samples of the remaining pixels in the sub-block window are tested using the two-sample KS hypothesis test method; Step S4: consider the pixel that has passed the test as a SHP point, and calculate the average intensity value E of all the SHP points. SHP , thereby estimating the background reflection intensity; Step S5: Using the SAR image after installing the CR, taking the pixel where the CR intensity peak is located in the SAR image as the center, selecting a target window of a predetermined size, and deducting the background intensity of the target window to obtain the true reflection intensity E of the CR. CR ; Step S6: Based on the average intensity value E SHP and the reflection intensity E CR The RCS value and the SCR value are calculated, and the temporal stability of the RCS value is analyzed to evaluate the stability of the CR state and exclude abnormal data.

2. The novel method for estimating scattering characteristics of a synthetic aperture radar corner reflector according to claim 1, characterized in that: in, Step S3 is to determine whether the time series sample and the reference sample are from the same distribution, thereby determining whether the time series sample pixel passes the test.

3. The novel method for estimating scattering characteristics of synthetic aperture radar corner reflectors according to claim 1 is characterized in that: in, The actual reflection intensity E of CR in step S5 CR The calculation formula is as follows: Where N represents the number of pixels contained in the window, is the backscattering intensity value of the window pixel, E SHP is the average intensity value.

4. The novel method for estimating scattering characteristics of a synthetic aperture radar corner reflector according to claim 1, characterized in that: in, The calculation formulas for the RCS value and the SCR value in step S6 are as follows: Where, ρ rg represents the orientation of the image, ρ az Represents the distance to pixel interval, θ loc is the local angle of incidence, The abnormal data includes water accumulation and damage of the CR.

Images