Different-satellite radiation cross calibration method based on natural point target RCS correction
By constructing the RCS theoretical model of natural point targets and the measurement-driven deviation correction model, the problem of insufficient accuracy in alien star cross calibration is solved, and high-precision alien star radiation cross calibration is achieved.
Patent Information
- Application Number
- CN202510910303.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-02
- Publication Date
- 2025-10-10
AI Technical Summary
The existing technology does not fully consider the RCS variation characteristics of natural point targets with incident angle and the differences in SAR system parameters in cross-satellite calibration, resulting in insufficient calibration accuracy.
The electromagnetic simulation method is used to construct a theoretical model of the isotropic RCS of natural point targets. The RCS values at different incident angles are fitted using the cosine power term model. A measurement-driven RCS deviation correction model is constructed to correct the RCS values at different satellites and incident angles. The response energy is calculated using the satellite images to be calibrated for radiation cross-calibration.
The accuracy of cross-calibration of inter-satellite radiation has been improved. Experiments show that the calibration accuracy has been improved by 3.7715dB, and the deviation of the calibration constant from the nominal value is only 0.994dB, achieving absolute cross-calibration of radiation between Sentinel-1A and GF-3.
Smart Images
Figure CN120761982A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of synthetic aperture radar radiation calibration, and in particular to a foreign star radiation cross-calibration method based on natural point target RCS correction. Background Art
[0002] Traditional Synthetic Aperture Radar (SAR) absolute radiometric calibration technology is based on artificial point targets such as corner reflectors and active calibrators. This requires the orderly arrangement of artificial calibrators at a specific calibration site and the implementation of synchronous ground measurements to accurately obtain attitude information such as pitch and azimuth angles, as well as position information such as latitude, longitude, and elevation.
[0003] Using natural point targets for absolute radiometric calibration can eliminate the burden of manual deployment and maintenance. Currently, natural point targets are primarily limited to absolute radiometric calibration within the same SAR satellite, making them difficult to meet the requirements for cross-calibration between satellites in practical applications. This is due to the following technical drawbacks: (1) The variation characteristics of the RCS of natural point targets with the incident angle are not fully considered. When the incident angles of the crossed binary satellites are greatly different, the RCS values of the natural point targets will have a large difference, resulting in insufficient calibration accuracy.
[0004] (2) The parameter differences of different SAR systems are not taken into account. There are differences in the SAR system parameters (such as radar center frequency, system noise, imaging time, etc.) between the calibrated satellite and the satellite to be calibrated. Directly using the target RCS value under the SAR system parameters of the calibrated satellite as the target RCS reference value under the SAR system parameters of the satellite to be calibrated will introduce additional RCS deviation, thereby affecting the calibration accuracy. Summary of the Invention
[0005] In order to solve the above technical problems, since the relevant technologies have not fully considered the characteristics of the RCS of natural point targets changing with the incident angle, this application uses electromagnetic simulation methods to perform theoretical modeling of the RCS of natural point targets in different directions under different incident angles to obtain the theoretical RCS values of natural point targets in all directions; according to the parameter differences of different SAR systems, a measurement-driven RCS deviation correction model is constructed to correct the theoretical RCS model in all directions, so as to realize the RCS value calculation considering different satellites, different incident angles and electromagnetic simulation fitting modeling deviations, and obtain the corrected RCS value of the satellite to be calibrated; secondly, the response energy of the natural point target is calculated using the image of the satellite to be calibrated, and finally the absolute radiation calibration constant of the satellite to be calibrated is calculated to complete the radiation cross-calibration between different satellites and improve the accuracy of the radiation cross-calibration between different satellites.
[0006] This application specifically proposes a cross-calibration method for alien radiation based on natural point target RCS correction, including the following steps: S1. Acquire dual-satellite synchronous observation data, and based on the target signal-to-clutter ratio (SCR) as the measurement standard, screen the calibrated satellite RCS value data slices and the uncalibrated satellite DN value data slices in the natural point target area; S2. Construct a theoretical model of the RCS of a natural point target in each direction under the satellite to be calibrated, obtain the theoretical RCS values of the natural point target in each direction under different incident angles, construct a measurement-driven RCS deviation correction model, and use the RCS value data slices of the calibrated satellite to perform corrections to obtain the target RCS value after RCS deviation correction; S3, calculating the response energy of the to-be-calibrated satellite DN value data slice obtained in step S1; S4. Perform radiation cross calibration using the target RCS value after RCS bias correction in step S2 and the target response energy of the satellite to be calibrated obtained in step S3 to obtain a final calibration constant.
[0007] In a preferred embodiment, in step S2, a theoretical model of the RCS of the natural point target under the satellite to be calibrated is constructed to obtain the theoretical RCS values of the natural point target under different incident angles. Specifically, a cosine power term model using the least squares fitting method is used to fit the theoretical RCS values of the natural point target under different incident angles. The expression of the theoretical RCS values of the natural point target under different incident angles is: ; ; Where, is the theoretical RCS value of each direction under different incident angles of natural point targets, in units of or dB, is the peak RCS value, is the angle of incidence, is the incident angle at which the peak RCS occurs, Indicates controlling the RCS attenuation rate.
[0008] In a preferred embodiment, the RCS theoretical value obtained by fitting is compared with the actual measured calibrated satellite RCS reference value to obtain the deviation variable , determine the target RCS value after RCS deviation correction; The target RCS value after RCS deviation correction is expressed as follows: ; Where, is the corrected RCS value of the satellite to be calibrated, in dB. is the deviation variable.
[0009] In a preferred embodiment, the deviation variable The calculation formula is: ; Where, is the RCS reference value of the calibrated satellite slice data, Indicates taking the average value.
[0010] In the preferred embodiment, in step S3, the response energy of the satellite DN value data slice to be calibrated is calculated by the integration method. : ; Where, represents the target integration area, is the number of pixels in the target integration area, is the background area, is the number of pixels in the background area, is the pixel number, is the pixel intensity value of the satellite to be calibrated, is the azimuth pixel value, is the distance pixel value.
[0011] In a preferred embodiment, in step S4, performing radiation cross calibration using the target RCS value after RCS bias correction in step S2 and the target response energy of the satellite to be calibrated obtained in step S3 to obtain the final calibration constant specifically includes: S4.1. Calculate the calibration constant of the satellite to be calibrated K : ; Where, is the RCS value of the natural point target after correction of the measured data; S4.2. Selection natural point targets as the target set, and calculate the average calibration constant according to the following formula ,Will As the final scaling constant: ; Where, The calculated A natural point target calibration constant.
[0012] In a preferred embodiment, in step S1, the natural point target area is screened by using the target signal-to-clutter ratio As screening criteria for natural point target areas: ; Where, is the point target peak impulse response power; is the average background power.
[0013] Screening out natural point targets as targets for cross calibration of radiation.
[0014] In the preferred embodiment, in the step S1, the acquiring of the dual-satellite synchronous observation data specifically comprises: performing screening on image synchronous observation data pairs by collecting imaging time, incident angle size, ascending / descending track information and polarization mode of the dual-satellite images; and performing data preprocessing on the screened image synchronous observation data pairs, including: track correction, coherent speckle filtering and geographic coding.
[0015] Compared with the prior art, the application has the following beneficial technical effects: (1) RCS model construction considering incident angle characteristics By establishing a three-dimensional model based on the physical structure parameters of the natural point target, the RCS values under different incident angles are simulated based on a geometric optics algorithm, and a RCS theoretical model in each direction is constructed by fitting with a cosine power term model, so as to realize the calculation of the RCS values of the natural point target under different incident angles. In the experimental verification, the RCS theoretical model in each direction of the natural point target under the condition of an incident angle of 30°-45° is obtained.
[0016] (2) RCS deviation correction model construction driven by actual measurement By comparing the RCS theoretical model in each direction with the actually measured RCS reference value of the calibrated satellite, a deviation variable is obtained, and a RCS deviation correction model driven by actual measurement is constructed, so as to solve the problem of low accuracy of cross calibration of radiation due to the use of only the RCS theoretical model in each direction. The cross calibration experiment of Sentinel-1A and GF-3 shows that the calibration accuracy is improved by an average of 3.7715 dB compared with the uncorrected method, and the average calibration constant of GF-3 obtained finally is 25.6826 dB, with a deviation of only 0.994 dB from the nominal value of GF-3.
[0017] (3) Cross calibration of radiation between different satellites The overall process of the application adopts an electromagnetic simulation RCS model in each direction corrected by actual measurement data, the RCS theoretical model in each direction is corrected by using the actual measurement RCS reference value of the calibrated satellite SAR, the RCS value under the SAR system parameter of the satellite to be calibrated is obtained, the response energy of the natural point target is calculated by using the image of the satellite to be calibrated, and finally the absolute cross calibration of radiation of the satellite to be calibrated is realized. The experiment shows that this scheme can realize the absolute cross calibration of radiation between Sentinel-1A and GF-3 by using the natural point target. BRIEF DESCRIPTION OF DRAWINGS
[0018] Figure 1 is a schematic diagram of the SAR absolute cross calibration of radiation based on the anisotropic deviation correction of the application; Figure 2 Detailed flow chart of alien radiation cross calibration for natural point target RCS correction in this application; Figure 3 Schematic diagram of the locations of offshore wind turbines screened for this application; Figure 4 Schematic diagram of the specific process of RCS electromagnetic simulation modeling for natural point targets in this application; Figure 5 Schematic diagram of the 3D model of the offshore wind turbine constructed for this application; Figure 6 Schematic diagram of the RCS theoretical model fitting for natural point targets constructed for this application; Figure 7 Schematic diagram of response energy calculation using the integration method for this application. DETAILED DESCRIPTION
[0019] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.
[0020] Example 1: S1: Acquire dual-satellite synchronous observation data, and based on the target signal-to-clutter ratio (SCR) as a measurement standard, screen the calibrated satellite RCS value data slices and the to-be-calibrated satellite DN value data slices of the natural point target area.
[0021] In this step, the target signal-to-noise ratio To measure the standard, natural point targets with strong scattering characteristics are screened out, and SAR image slice data of natural point targets are obtained for subsequent cross-calibration.
[0022] S11: Acquire binary star synchronous observation data.
[0023] This step determines the time range and geographical area required for the study, selects an appropriate satellite imaging area, and screens out simultaneous observation data pairs from two different SAR satellites. By collecting detailed parameters such as imaging time, incident angle, ascending and descending orbit information, and polarization mode of the binary images, the simultaneous observation data pairs are screened according to the following requirements: The time difference should be ≤15 days to ensure the consistency of imaging climate conditions; For the same ascending and descending orbits, the azimuth angle difference should be less than 5°; Same polarization mode.
[0024] After screening out the synchronous observation data pairs of L1A-level image products, they are subjected to data preprocessing, including orbit correction, speckle filtering, geocoding, etc., in preparation for subsequent operations.
[0025] S12: Based on signal-to-noise ratio Screening natural point targets with strong scattering characteristics.
[0026] In this step, the signal-to-clutter ratio is used to calculate the natural point targets in the imaging area. As a natural point target screening criterion, calculate the target signal-to-clutter ratio The formula is as follows: ; Where, is the point target peak impulse response power; is the average background power.
[0027] when When the background noise has little effect on the calibration results (it can be considered that the effect on the calibration results is less than 0.5dB), the calibration accuracy can be effectively guaranteed. The natural point target is used as the target for subsequent radiation cross calibration, and the area containing the concentrated natural point target is sliced to obtain the calibrated satellite RCS value data slices and the to-be-calibrated satellite DN value data slices in the area.
[0028] S2: Construct a theoretical model of the RCS of the natural point target in each direction under the satellite to be calibrated, obtain the theoretical RCS values of the natural point target in each direction under different incident angles, construct an RCS deviation correction model driven by field measurements, and use the RCS value data slices of the calibrated satellite to perform corrections to obtain the target RCS value after RCS deviation correction.
[0029] This step uses modeling to correct the RCS of the filtered data. There is a deviation in the RCS of natural point targets between the calibrated satellite and the calibrated satellite. This deviation can be divided into two aspects: the first is the deviation caused by the difference in incidence angle, which will be solved by the electromagnetic simulation RCS modeling method; the second is the deviation caused by different SAR satellite systems and electromagnetic simulation modeling, which will be calculated and corrected by introducing a fixed deviation. The details are as follows: S21: Geometric modeling of natural point targets and construction of electromagnetic simulation models.
[0030] This step first collects and filters physical structural data of the natural point target, including its dimensions, three-dimensional views, and multi-angle images. A 3D model is then constructed using 3D graphics software to ensure its accuracy and reliability. The constructed 3D model is then imported into electromagnetic simulation software, where electromagnetic parameters, including radar center frequency and polarization, are set based on the characteristics of both the calibrated and uncalibrated satellites. Finally, a fine mesh is created, and an algorithm based on geometric optics is used to simulate the RCS values of the natural point target at different incident angles. This results in a theoretical model of the RCS of the natural point target in all directions under the uncalibrated satellite.
[0031] S22: Isotropic RCS theoretical model fitting.
[0032] This step uses a fitting method to describe the RCS of natural point targets at different incident angles. Considering that the RCS of natural point targets is usually concentrated in the main lobe as the incident angle changes, and the change with the incident angle will be similar to the cosine function or its power deformation, this application uses a cosine power term model to fit the simulation data.
[0033] Specifically, the cosine power term model using the least squares method is used to fit the theoretical RCS values of the natural point target at different incident angles. The expression of the theoretical RCS value of each direction is: ; ; Where, is the theoretical RCS value of each direction under different incident angles of natural point targets, in units of or dB, is the peak RCS value, is the angle of incidence, is the incident angle at which the peak RCS occurs, in degrees, The value is obtained by regression fitting of simulation data, which controls the RCS attenuation rate. The larger the value, the faster the RCS changes with the incident angle.
[0034] S23: Alien SAR system parameter deviation correction. This step builds a measurement-driven RCS deviation correction model and introduces deviation variables. Indicates the deviation between different SAR satellite system parameters except the incident angle and the deviation caused by electromagnetic simulation modeling. Compare the fitted RCS theoretical value with the actual measured RCS reference value of the calibrated satellite to obtain the deviation variable , determine the target RCS model after RCS deviation correction; The target RCS model expression after RCS deviation correction is as follows: ; Where, is the corrected RCS value of the satellite to be calibrated, in dB. is the theoretical model of the isotropic RCS obtained above, is the deviation variable, which represents the fixed deviation between the RCS theoretical model and the RCS reference value in each direction.
[0035] Among them, the deviation variable The calculation formula is: ; Where, is the RCS reference value of the calibrated satellite slice data obtained above, is the theoretical RCS value obtained from the isotropic RCS theoretical model, Indicates taking the average value.
[0036] S3: Calculate the response energy of the to-be-calibrated satellite DN value data slice obtained in step S1.
[0037] This step uses the integration method and other methods to calculate the response energy of the natural point targets that have been screened and modeled and simulated, and the data slices of the satellite DN value to be calibrated obtained in S1. The calculation formula of the integration method is as follows: ; Where, represents the target integration area, is the number of pixels in the target integration area, is the background area, is the number of pixels in the background area, is the pixel intensity value of the satellite to be calibrated, is the azimuth pixel value, is the distance pixel value, the integral method calculation diagram is as follows Figure 7 .
[0038] S4: Perform radiation cross calibration using the target RCS value after RCS bias correction in step S2 and the target response energy of the satellite to be calibrated obtained in step S3 to obtain the final calibration constant.
[0039] S4.1. Calculate the calibration constant of the satellite to be calibrated K。
[0040] This step uses the target RCS value after RCS bias correction in S2 and the target response energy of the satellite to be calibrated obtained in S3 to perform radiation cross calibration. The calibration constant of the satellite to be calibrated is calculated according to the following formula: K : ; Where, is the response energy of the natural point target corresponding to the satellite to be calibrated, is the RCS value of the natural point target after correction of the measured data.
[0041] S4.2. Selection natural point targets as the target set, and calculate the average calibration constant according to the following formula ,Will as the final calibration constant.
[0042] In order to improve the calibration accuracy, this application selects natural point targets as the target set, and calculate the average calibration constant according to the following formula As the final scaling constant: ; Where, The calculated A natural point target calibration constant.
[0043] Example 2: This example uses offshore wind turbines as representative natural point targets for the experiment. Offshore wind turbines have the following characteristics: large-scale offshore wind farm construction provides a natural calibration target network, providing quantitative support for the selection of natural point targets. Furthermore, offshore wind turbines are evenly distributed, with low and stable interference between turbines, facilitating calibration. The scattering interference from the open sea surface is significantly lower than that from the land environment, providing excellent conditions for extracting the scattering characteristics of natural point targets.
[0044] This embodiment proposes Figure 1 The SAR radiometric cross-calibration method based on anisotropy correction is shown in FIG. 1 ; specifically, the Sentinel-1A and GF-3 radiometric cross-calibration process based on natural point target RCS correction is shown in FIG. Figure 2 shown.
[0045] The specific process of this embodiment includes: S1. Obtain binary star synchronous observation data based on the target signal-to-noise ratio To measure the standard, the calibrated satellite RCS value data slices and the to-be-calibrated satellite DN value data slices of the natural point target area are screened.
[0046] This step mainly uses the target signal-to-noise ratio In order to measure the standards, offshore wind turbine targets with strong scattering characteristics are screened out, and offshore wind turbine targets and corresponding SAR satellite image slice data for subsequent cross-calibration are obtained.
[0047] S11: Acquire binary star synchronous observation data.
[0048] This step determines that the study area focuses on the offshore wind farm near the Shanghai Donghai Bridge. The study period is from December 2023 to July 2024. The calibrated satellite is Sentinel-1A (hereinafter referred to as S1A), and the satellite to be calibrated is GF-3. GF-3 wide-swath fine-swath (FSII) data for the study area and study period are obtained, and S1A interferometric wide-swath (IW) data for the corresponding area and time are searched from the ESA official website. The image synchronization observation data pairs are selected according to the following requirements: The time difference should be ≤15 days, and the sea and climate conditions should be consistent; For the same ascending and descending orbits, the azimuth angle difference should be less than 5°; Same polarization mode.
[0049] Finally, three sets of binary-satellite synchronous observation data pairs were screened, as shown in Table 2. Professional image processing software was used to convert the acquired data format, including pre-processing processes such as orbit correction, speckle filtering, and geocoding, and finally converted to a universal image format. For the calibrated satellite S1A data, the Refined Lee method was used to remove the filter in the image, and the filter window size was set to 7×7 pixels to improve the image quality. At the same time, the calibrated satellite also needs to be calibrated by radiation to obtain the calibrated backscatter coefficient RCS value. For the GF-3 data of the uncalibrated satellite, the Frost method was used for filtering, and the filter window size was set to 5×5. Both used the WGS84 coordinate system and a resolution of 10m, and used the satellite orbit parameters and ground control point information to perform geo-coordinate correction on the images to ensure the accurate geo-positioning of the images.
[0050] Table 2 Detailed information of the selected binary star synchronous observation data pairs
[0051] S12: Based on signal-to-noise ratio Screening natural point targets with strong scattering characteristics.
[0052] In this embodiment, the natural point target with strong scattering characteristics is an offshore wind turbine.
[0053] In an offshore wind farm near the Shanghai Donghai Bridge, the peak impulse response power of a certain offshore wind turbine was obtained by analyzing its pixel intensity in the SAR image. , and combined with the background area pixel statistical information to obtain the background average power , and then calculate according to the formula Value, filter out The offshore wind turbines were selected as the target points for subsequent cross calibration. Finally, this experiment selected 39 offshore wind turbines as the target points for radiation cross calibration, such as Figure 3As shown in the figure, the red box is the selected offshore wind turbine target point.
[0054] S2. Construct a theoretical model of the RCS of a natural point target in each direction under the satellite to be calibrated, obtain the theoretical RCS values of the natural point target in each direction under different incident angles, construct an RCS deviation correction model driven by field measurements, and use the RCS value data slices of the calibrated satellite to perform corrections to obtain the target RCS value after RCS deviation correction.
[0055] This step uses modeling to correct the RCS of the filtered data. The following measures are taken to address the RCS deviation between the satellite to be calibrated and the calibrated satellite: First, electromagnetic simulation is used to construct an isotropic RCS theoretical model to obtain the RCS at different incident angles and correct the deviation caused by the difference in incident angle in the imaging geometry of the calibrated and uncalibrated satellites; second, a fixed deviation is introduced to quantify the deviation caused by different SAR satellite systems and electromagnetic simulation modeling deviations.
[0056] S21: Geometric modeling of natural point targets and construction of electromagnetic simulation models.
[0057] In this embodiment, the natural point target is an offshore wind turbine, and geometric modeling and electromagnetic simulation modeling of the natural point target are performed.
[0058] The steps of electromagnetic simulation modeling of offshore wind turbines are as follows: Figure 4 First, we collected and screened the physical structure data of offshore wind turbines, including dimensions, three-view drawings, and multi-angle images. We used Blender software to build a 3D model of an offshore wind turbine with an actual blade length of 64 meters and a tower height of 90 meters. We then drew the geometric shapes of the blades and tower according to the actual situation, and set appropriate material properties to simulate the actual electromagnetic reflection characteristics. The three-view drawings of the constructed offshore wind turbine model are shown below. Figure 5 As shown, (a) front view, (b) side view, and (c) top view.
[0059] The constructed three-dimensional model was then imported into electromagnetic simulation software, and simulations were performed on the S1A and GF-3 satellites at a frequency of 5.4 GHz and an incidence angle range of 30°-45°. Fine meshing was performed based on an adaptive meshing algorithm, and the ray launching geometrical optics method (RL-GO) was used to simulate the RCS values of offshore wind turbines under different conditions.
[0060] S22: Isotropic RCS theoretical model fitting.
[0061] This step uses fitting to describe the RCS of offshore wind turbines at different incident angles. Since the RCS of offshore wind turbines changes with the incident angle and is similar to a cosine function or its power deformation, a cosine power term model is used to fit the simulation data and construct an isotropic RCS theoretical model. The fitting model is as follows: Figure 6 As shown, the cosine power term model fitting result formula is as follows: ; Where, is the theoretical RCS value of offshore wind turbines in all directions, in dB. is the incident angle, in degrees, and the peak RCS value The incident angle at which the peak RCS occurs is 28.82dB. is 39.47°, The value is 120.89.
[0062] S23: Alien SAR system parameter deviation correction.
[0063] This step builds the RCS deviation correction model driven by field measurement and introduces the deviation variable Indicates the deviation between different SAR satellite system parameters except the incident angle and the deviation caused by electromagnetic simulation modeling. Compare the fitted RCS theoretical model with the actual measured calibrated satellite RCS reference value to obtain the deviation variable , determine the final bias-corrected RCS model, the bias variables in the RCS bias-corrected model The solution results are as follows: ; The final target RCS model after RCS deviation correction is: ; S3. Calculation of offshore wind turbine response energy.
[0064] This step mainly targets offshore wind turbines that have been screened and modeled, and calculates their response energy through the integration method for subsequent calculation of calibration constants.
[0065] S3: Calculate the response energy of the to-be-calibrated satellite DN value data slice obtained in step S1.
[0066] In this embodiment, for the offshore wind turbines that have been screened and modeled, the response energy is calculated using the integration method for the data slice of the satellite DN value to be calibrated obtained in S1. The schematic diagram of the integration method is as follows: Figure 7 The final average response energy of offshore wind turbines in the uncalibrated SAR images from each dual-satellite synchronous observation data pair is shown in Table 3.
[0067] Table 3 Average response energy of target points in the SAR image to be calibrated
[0068] S4: Perform radiation cross calibration using the target RCS value after RCS bias correction in step S2 and the target response energy of the satellite to be calibrated obtained in step S3 to obtain the final calibration constant.
[0069] S4.1. Calculate the calibration constant of the satellite to be calibrated K。
[0070] This step uses the target RCS value after RCS bias correction in S2 and the target response energy of the satellite to be calibrated obtained in S3 to perform radiation cross calibration. The calibration constant of the satellite to be calibrated is calculated according to the following formula: K : ; Where, is the response energy of the natural point target corresponding to the satellite to be calibrated, is the RCS value of the natural point target after correction of the measured data.
[0071] S4.2. Selection natural point targets as the target set, and calculate the average calibration constant according to the following formula ,Will as the final calibration constant.
[0072] In this embodiment, in order to improve the calibration accuracy, 39 offshore wind turbines are selected for radiation cross calibration, and the average calibration constant is calculated according to the formula As the final average scaling constant (dB): ; The calibration constant obtained by the method of this embodiment is compared with the nominal calibration constant and the calibration constant without RCS deviation correction. The results and comparison are shown in Table 4.
[0073] Table 4 Results and comparison of this example
[0074] The final average calibration constant of the GF-3 satellite to be calibrated is 25.6826dB, which is consistent with the nominal calibration constant of GF-3. K Compared with the deviation of 0.994dB, the calibration accuracy is improved by 3.7715dB compared with the method without RCS deviation correction, which verifies the effectiveness of the alien star radiation cross-calibration method based on natural point target RCS correction proposed in this application, and provides a new idea for the radiation consistency test of multi-star system.
[0075] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.
Claims
1. A cross-calibration method for alien radiation based on natural point target RCS correction, characterized in that: The steps include: S1. Acquire dual-satellite synchronous observation data, and based on the target signal-to-clutter ratio (SCR) as the measurement standard, screen the calibrated satellite RCS value data slices and the uncalibrated satellite DN value data slices in the natural point target area; S2. Construct a theoretical model of the RCS of a natural point target in each direction under the satellite to be calibrated, obtain the theoretical RCS values of the natural point target in each direction under different incident angles, construct a measurement-driven RCS deviation correction model, and use the RCS value data slices of the calibrated satellite to perform corrections to obtain the target RCS value after RCS deviation correction; S3, calculating the response energy of the to-be-calibrated satellite DN value data slice obtained in step S1; S4. Perform radiation cross calibration using the target RCS value after RCS bias correction in step S2 and the target response energy of the satellite to be calibrated obtained in step S3 to obtain a final calibration constant.
2. The alien star radiation cross calibration method based on natural point target RCS correction according to claim 1 is characterized in that: In step S2, a theoretical model of the RCS of the natural point target under the satellite to be calibrated is constructed to obtain the theoretical RCS values of the natural point target under different incident angles. Specifically, a cosine power term model using the least squares fitting method is used to fit the theoretical RCS values of the natural point target under different incident angles. The expression of the theoretical RCS values of the natural point target under different incident angles is: ； ； Where, is the theoretical RCS value of each direction under different incident angles of natural point targets, in units of or dB, is the peak RCS value, is the angle of incidence, is the incident angle at which the peak RCS occurs, Indicates controlling the RCS attenuation rate.
3. The alien star radiation cross calibration method based on natural point target RCS correction according to claim 2 is characterized in that: Compare the fitted RCS theoretical values with the actual measured RCS reference values of the calibrated satellites to obtain the deviation variables , determine the target RCS value after RCS deviation correction; The target RCS value after RCS deviation correction is expressed as follows: ； Where, is the corrected RCS value of the satellite to be calibrated, in dB. is the deviation variable.
4. The alien star radiation cross calibration method based on natural point target RCS correction according to claim 3 is characterized in that: Deviation variables The calculation formula is: ； Where, is the RCS reference value of the calibrated satellite slice data, Indicates taking the average value.
5. The alien star radiation cross calibration method based on natural point target RCS correction according to claim 1 is characterized in that: In step S3, the response energy of the satellite DN value data slice to be calibrated is calculated by the integration method. : ； Where, represents the target integration area, is the number of pixels in the target integration area, is the background area, is the number of pixels in the background area, is the pixel number, is the pixel intensity value of the satellite to be calibrated, is the azimuth pixel value, is the distance pixel value.
6. The alien star radiation cross calibration method based on natural point target RCS correction according to claim 5 is characterized in that: In step S4, performing radiation cross calibration using the target RCS value after RCS bias correction in step S2 and the target response energy of the satellite to be calibrated obtained in step S3 to obtain the final calibration constant specifically includes: S4.
1. Calculate the calibration constant of the satellite to be calibrated K : ； Where, is the RCS value of the natural point target after correction of the measured data; S4.
2. Selection natural point targets as the target set, and calculate the average calibration constant according to the following formula ,Will As the final scaling constant: ； Where, The calculated A natural point target calibration constant.
7. The alien star radiation cross calibration method based on natural point target RCS correction according to claim 1 is characterized in that: In step S1, the natural point target area is screened by using the target signal-to-clutter ratio to select the natural point target area in the imaging area. As screening criteria for natural point target areas: ； Where, is the peak impulse response power of the point target; is the background average power; Filter out The natural point target is used as the target of radiation cross calibration.
8. The alien star radiation cross calibration method based on natural point target RCS correction according to claim 1 is characterized in that: In step S1, obtaining the binary star synchronous observation data specifically includes: screening the image synchronous observation data pairs by collecting the imaging time, incident angle, ascending and descending orbit information, and polarization mode of the binary star images; and performing data preprocessing on the screened image synchronous observation data pairs, including orbit correction, coherent speckle filtering, and geocoding.