A method for extracting significant wave height based on synthetic aperture radar image

Abstract

The application provides a significant wave height extraction method based on a synthetic aperture radar image, step 1: data acquisition, matching the altimeter significant wave height with the synthetic aperture radar interferometric imaging mode data to obtain a sample data set. Step 2: the data amount of the altimeter significant wave height in different incidence angle ranges is counted to obtain the numerical distribution of the altimeter significant wave height in the global near-shore range. Step 3: the normalized variance of the backscattering coefficient of the synthetic aperture radar and the altimeter significant wave height are subjected to quality control to obtain the significant wave height and normalized variance data. Step 4: a linear relationship between the significant wave height and the normalized variance is established. Step 5: the estimated value of the synthetic aperture radar significant wave height is evaluated and verified, and the significant wave height can be obtained by inputting only a single image parameter, the normalized variance. The application not only considers the synthetic aperture radar open ocean wave mode, but also relates to the near-shore research, and provides technical support for the development of related topics in the coastal area.

Description

Technical Field

[0001] The present invention relates to a method for processing ocean data, and in particular to a method for extracting effective wave height based on synthetic aperture radar images. Background Art

[0002] Significant wave height (SWH) is defined as the average height of one-third of the highest wave heights in a given wave train. my country is a major maritime nation, with approximately 23%-37% of the world's population living within 100 kilometers of the coastline. Measuring wave height in coastal areas plays a vital role in coastal protection, coastal safety, and coastal risk assessment. With the rapid development of the marine economy, human activities at sea are increasing, and ships and various marine industries require information on wave heights for route planning. Furthermore, natural disasters pose a significant threat to marine operations. The occurrence of natural disasters such as typhoons and storm surges inevitably causes changes in marine wave parameters. Research and application of wave height measurement and sea condition forecasting can effectively improve humanity's ability to prevent natural disasters and reduce national economic losses under extreme disaster conditions. Ocean surface wave measurements are valuable in various fields, and significant wave height is one of the most important parameters in wave observation, making its measurement essential.

[0003] Early measurements of significant wave height were made using in-situ measurement stations, including buoys. However, these stations were limited in their distribution, leaving many localized fluctuations unmeasured. Satellite altimetry effectively addressed this shortcoming. Its measurement principle relies on estimating the propagation time of a radar pulse from the satellite to the ground and back, allowing the significant wave height to be estimated based on the resulting waveform. However, satellite altimetry still has limitations and is only applicable to spatially uniform areas, such as the open ocean. Accurately analyzing the significant wave height over an uneven sea surface is difficult.

[0004] With the development of satellite radar altimeters, the emergence of time-delay Doppler altimeters, and the launch of the first synthetic aperture radar altimeter, CryoSat-2, a new era of ocean remote sensing has begun. Raney (1998) proposed the concept of time-delay Doppler altimeters, incorporating the concept of aperture synthesis into altimeters. This has significantly increased along-track resolution from several kilometers to approximately 300 meters, significantly improving altitude accuracy. Because synthetic aperture radar altimeters offer twice the accuracy of traditional altimeters, they will play a vital role in mesoscale ocean measurements, ocean depth exploration, and high-precision ocean gravity field measurements. Major space powers have launched numerous radar altimeter satellites.

[0005] The technology for applying satellite radar altimeters is now quite mature. Janssen used a triple matching method to demonstrate the high accuracy of Envisat altimeter data. Li and Holt used buoys and Envisat to verify the accuracy of North Atlantic and European wave models. The significant wave heights of the North Atlantic and European empirical models were basically consistent with the altimeter and buoy observations. Park used buoys in the waters around South Korea to demonstrate that the accuracy errors of the significant wave height data from Topex / Poseidon, Envisat, Jason-1, and Jason-2 were very small. Therefore, high-precision satellite radar altimeters and the National Data Buoy Center (NDBC) of the National Oceanic and Atmospheric Administration (NOAA) are often used as independent data sources to evaluate and verify the accuracy of other methods for obtaining significant wave heights.

[0006] Space-borne synthetic aperture radar (SAR) is a powerful tool for detecting sea state from significant wave height. Compared with traditional in-situ observation of sea waves, SAR can provide high spatial resolution sea wave spectrum all day and all weather, which shows unique ability and potential for sea wave observation. The ocean surface is determined by sea state, and the basic parameter of sea state is significant wave height. Therefore, it is of great significance to use SAR measured data to retrieve the sea wave inversion model of significant wave height. The synthetic aperture radar significant wave height inversion scheme is divided into two categories. The first method is to retrieve the sea wave spectrum from SAR image spectrum to obtain the significant wave height, such as the Max-Planck Institute model, semi-parametric inversion model, etc. However, due to the influence of SAR's special imaging mechanism of velocity bunching, SAR images often appear very blurred in the azimuth direction, and the derived wave spectrum is prone to distortion. Due to the random motion of the sea waves, the Doppler frequency shift of the synthetic aperture radar backscattering echo signal will occur, and the process of sea wave imaging will show strong nonlinearity, which will affect SAR imaging. Therefore, only by introducing other numerical model results as prior information or through SAR image cross spectrum can we obtain the complete sea wave spectrum. Among them, the cross spectrum calculation process of synthetic aperture radar image does not need initial guess information, which solves the problem of 180° ambiguity of sea wave propagation direction, but the obtained sea wave spectrum cannot analyze the information of high-frequency short waves. It can be seen that the method of retrieving significant wave height from SAR image spectrum still has limitations. The second method is an empirical model, which can directly extract significant wave height from SAR image, and no longer depends on sea wave spectrum to calculate sea wave parameters, avoiding the complex inversion process. At first, Schulz-Stellenfleth proposed the CWAVE model, which can extract radar cross section, variance and other parameters from SAR image, and directly establish the relationship between these parameters and significant wave height to obtain significant wave height. On the basis of CWAVE model, Schulz-Stellenfleth et al. established the CWAVE_ERS based on ERS2 data. Li and Stopa et al. proposed the CWAVE_ENV model based on ENVISAT ASAR data and the CWAVE-S1A model based on Sentinel-1A data on the basis of CWAVE-ERS model. In addition, with the continuous progress of sea wave parameter inversion research, people gradually realize that there is a strong correlation between the azimuth cutoff wavelength and the significant wave height, which can be used for significant wave height inversion research. The azimuth cutoff wavelength is the smallest wavelength that can be observed along the track direction, which can be used as a measure of the azimuth resolution of synthetic aperture radar. Therefore, Stopa et al. used the strong dependence between the azimuth cutoff wavelength and the significant wave height to establish the CWAVE_Fnn model based on Sentinel-1 data through neural network algorithm.Because these empirical models do not involve complex transfer modulation functions, they can directly establish the relationship between SAR image spectra and significant wave heights, resulting in relatively high inversion accuracy. However, these empirical models were developed and validated for open ocean wave patterns and have not yet been studied for near-coastal interferometric imaging, resulting in limitations in marine environments. Summary of the Invention

[0007] 1. Technical problems to be solved:

[0008] How to establish an empirical model to obtain significant wave height in complex ocean environments. Empirical models usually require multiple input parameters to obtain significant wave height. The algorithm is computationally intensive and the processing is complex, making it difficult to invert significant wave height on a global scale.

[0009] 2. Technical solution:

[0010] In order to solve the above problems, the present invention provides a method for extracting effective wave height based on synthetic aperture radar images, comprising the following steps:

[0011] Step 1: Obtain the data of synthetic aperture radar interferometry imaging mode and altimeter effective wave height in the sea area studied for many years, match the altimeter effective wave height with the synthetic aperture radar interferometry imaging mode data, set the appropriate time matching window and spatial matching distance, and obtain the matched sample data set.

[0012] Step 2: Select the altimeter significant wave height data of the sample data set, analyze the global distribution of the altimeter significant wave height, divide it into different incidence angle ranges, count the data volume of the altimeter significant wave height in different incidence angle ranges, and obtain the numerical distribution of the altimeter significant wave height in the global near-coast range.

[0013] Step 3: Perform quality control on the normalized variance of the synthetic aperture radar backscatter coefficient and the altimeter effective wave height of the sample data set. Use the confidence ellipse fitting method, set the quality control conditions for confidence ellipse fitting, and obtain the fitted effective wave height and normalized variance data.

[0014] Step 4: Perform linear fitting on the fitted significant wave height and normalized variance data at different incident angles to obtain the fitting coefficients at each incident angle. Establish a linear relationship between significant wave height and normalized variance, input the normalized variance, and obtain the estimated value of the synthetic aperture radar significant wave height through the model.

[0015] Step 5: Use the altimeter and buoy significant wave height data to evaluate and validate the SAR significant wave height estimate. Compare the average deviation of the SAR significant wave height estimate compared to the altimeter significant wave height at each incidence angle. Calculate the deviation and standard deviation of the SAR significant wave height estimate compared to the buoy significant wave height.

[0016] In step 1, the altimeters used include Cryosat-2 and Sentinel-3_a launched by the European Space Agency, Jason-2 launched by the Delta-2 carrier rocket, Jason-3 launched by the Falcon 9 carrier rocket, and SARAL launched by the French National Center for Space Studies and the Indian Space Research Institute. Their resolution is 6 km, and significant wave height data near the coast of the world from 2019 to 2022 are selected.

[0017] In step 1, the synthetic aperture radar used is the interferometric imaging mode of the Sentinel-1 in the European Space Agency's Copernicus program, with a resolution of 10m and a synthetic aperture radar image size of 250kmx250km.

[0018] The time matching window and spatial matching distance are as follows: in terms of time, ±1.5 hours is used as the time matching window; in terms of space, the data points on the synthetic aperture radar image are matched within a circular range with a radius of 3 km and the center of the circle obtained by the altimeter as the center of the circle with a radius of 3 km. The altimeter circular range with a synthetic aperture radar image pixel less than 500 is eliminated, and the matching is completed.

[0019] In step 2, the sample significant wave height data are distributed in the near-coast areas of the world, with the incident angle ranges of 32°±1°, 34°±1°, 36°±1°, 38°±1°, and 40°±1°. The sample data sets exceed 67,000, of which more than 12,000 are within the 38°±1° incident angle range, and 6,000±100 are within the 32°±1° incident angle range. In the near-coast area of ​​the world, the significant wave heights are mainly concentrated around 2m±0.5, with more than 5,000 data points, showing a Weibull distribution.

[0020] The global near coasts include the coast of Hawaii, the east and west coasts of the United States, the west coast of Europe, the southeast coast of Africa, the southeast coast of China and the coast of Japan.

[0021] In step 3, the quality control is performed on all data points using a confidence ellipse fitting method, where the normalized variance consists of the mean and the standard deviation, and the formula is as follows:

[0022]

[0023] Where σ is the standard deviation, μ is the mean, and Nvar is the normalized variance.

[0024] In step 4, the linear fitting is a one-variable first-order polynomial fitting, and the calculation formula is:

[0025] y=ax+b

[0026] Where y is the estimated value of the synthetic aperture radar effective wave height, x is the normalized variance, and a and b are the fitting coefficients of the polynomial.

[0027] In step 5, the buoy significant wave height is based on the significant wave height data of the global coastal waters from 2019 to 2022 from the National Data Buoy Center of the National Oceanic and Atmospheric Administration of the United States.

[0028] In step 5, the specific method includes the following steps:

[0029] Step 101: Use the altimeter significant wave height to evaluate the estimated SAR significant wave height. Compare the average deviation of the SAR significant wave height estimate at each incident angle compared to the altimeter significant wave height, also known as the weighted deviation. The calculation formula is:

[0030]

[0031] Among them, w1, w2, ... ..., w n are the weights of the synthetic aperture radar effective wave height estimation value, x1, x2, ... ... , x n is the median value of the synthetic aperture radar effective wave height estimate for each 0.2m interval, is the weight bias of the synthetic aperture radar significant wave height estimate;

[0032] Step 102: Verify the estimated SAR effective wave height using the on-site observed effective wave height data, and calculate the deviation and standard deviation of the estimated SAR effective wave height compared to the on-site observed effective wave height.

[0033] The deviation is calculated as follows:

[0034] bias(x)=yf(x)#(4)

[0035] Where y is the estimated value of the SAR effective wave height, f(x) is the effective wave height observed on site, and bias(x) is the deviation of the estimated value of the SAR effective wave height compared to the effective wave height observed on site.

[0036] The formula for calculating the standard deviation is as follows:

[0037]

[0038] Where X is a single SAR effective wave height estimate, μ is the mean of the SAR effective wave height estimates, N is the total number of samples of the SAR effective wave height estimates, and STD is the standard deviation of the SAR effective wave height estimates.

[0039] 3.Beneficial effects:

[0040] The present invention uses only the normalized variance of a single image parameter as input to obtain significant wave heights through an empirical model. Compared to traditional methods, this method does not rely on multiple input parameters, improving the simplicity of model calculations. This method not only considers synthetic aperture radar open ocean wave patterns but also addresses near-coastal research, providing technical support for related research in coastal areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a technical flow chart of extracting significant wave height based on synthetic aperture radar images in the present invention;

[0042] Figure 2 Figure 3: Quality control of significant wave height and normalized variance and establishment of empirical model. (a) shows the confidence ellipse fitting and linear fitting at an incident angle of 36°±1°, and (b) shows the comparison of linear fitting results at all incident angles. DETAILED DESCRIPTION

[0043] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0044] The present invention provides a method for extracting effective wave height from synthetic aperture radar images in near-coast interferometric imaging mode. An empirical model is established based on a single parameter, and only the normalized variance is input to obtain the SAR effective wave height estimation value. Figure 1 As shown, the method specifically includes the following steps:

[0045] Step 1: Obtain the data of synthetic aperture radar interferometry imaging mode and altimeter effective wave height in the sea area studied for many years, match the altimeter effective wave height with the synthetic aperture radar interferometry imaging mode data, set the appropriate time matching window and spatial matching distance, and obtain the matched sample data set.

[0046] In one embodiment, there are five altimeters, namely Cryosat-2 and Sentinel-3_a launched by the European Space Agency, Jason-2 launched by the Delta-2 carrier rocket, Jason-3 launched by the Falcon 9 carrier rocket, and SARAL launched by the French National Center for Space Studies and the Indian Space Research Institute. The spatial resolution is 6 km, and the significant wave height data of the global nearshore waters from 2019 to 2022 are selected.

[0047] In one embodiment, the synthetic aperture radar used is Sentinel-1 in the Copernicus program of the European Space Agency, which operates in an interferometric imaging mode with a spatial resolution of 10 m and a SAR image size of 250 km x 250 km.

[0048] Due to the different spatial resolutions of the altimeter and synthetic aperture radar, spatiotemporal data matching was required. Temporally, a ±1.5-hour time window was used. Spatially, a circular range with a radius of 3 km, centered at the center of the altimeter's significant wave height, was used to match data points on the synthetic aperture radar image. Any areas within the altimeter's circular range with fewer than 500 synthetic aperture radar pixels were excluded, and all matching data points were used as the sample dataset for model construction.

[0049] Step 2: Select the altimeter significant wave height data of the sample data set, analyze the global distribution of the altimeter significant wave height, divide it into different incidence angle ranges, count the data volume of the altimeter significant wave height in different incidence angle ranges, and study the numerical distribution of the altimeter significant wave height on a global scale.

[0050] The sample significant wave height data are distributed along the global coast. In one embodiment, the global coast includes the coast of Hawaii, the east and west coasts of the United States, the west coast of Europe, the southeast coast of Africa, the southeast coast of China, and the coast of Japan.

[0051] In one embodiment, the sample dataset contains over 67,000 data points, covering incident angles of 32°±1°, 34°±1°, 36°±1°, 38°±1°, and 40°±1°. The 38°±1° incident angle range contains the most data, exceeding 12,000 points, while the 32°±1° incident angle range contains the least data, approximately 6,000 points. The data for each incident angle range is sufficient to support research. Globally, significant wave heights near coastal areas are primarily concentrated around 2 meters, with over 5,000 points exhibiting a Weibull distribution.

[0052] Step 3: Perform quality control on the normalized variance of the synthetic aperture radar backscatter coefficient and the altimeter effective wave height of the sample data set. Use the confidence ellipse fitting method, set the quality control conditions for confidence ellipse fitting, and obtain the fitted effective wave height and normalized variance data.

[0053] The mean and standard deviation of the backscatter coefficient of a synthetic aperture radar are two of the most commonly used parameters for describing the statistical characteristics of a radar image. From a statistical perspective, they represent the first-order moment and second-order moment of the radar echo, respectively. The combination of these two variables is called the normalized variance, and its formula is as follows:

[0054]

[0055] Where σ is the standard deviation, μ is the mean, and Nvar is the normalized variance.

[0056] Nvar and SWH quality control of sample data points, in an embodiment, confidence ellipse fitting is used, the set confidence ellipse standard deviation is 1.5, the screening condition is that the range of effective wave height is [0, 6], the range of normalized variance is less than 0.15, and the range of incidence angle is 36°±1°. As shown in Figure 2 (a), the obtained red ellipse is the fitted confidence ellipse, the short axis length is 0.07, the long axis length is 2.95, the center is (0.07, 1.99), and the central rotation angle is 179.4°.

[0057] Step 4: Linear fitting is performed on the fitted effective wave height and normalized variance data under different incidence angle ranges to obtain the fitting coefficients under each incidence angle, and a linear relationship between the effective wave height and the normalized variance is established, and the synthetic aperture radar effective wave height estimation value can be obtained through the model by only inputting the normalized variance.

[0058] The linear fitting formula is as follows:

[0059] y=ax+b

[0060] Wherein, y is the SAR effective wave height estimation value, x is the normalized variance, and a and b are the fitting coefficients of the polynomial. In an embodiment, the fitting coefficients a corresponding to the above incidence angle range are 41.87, 40.49, 40.71, 41.57, and 42.39 respectively, and the fitting coefficients b are -1.11, -0.87, -0.79, -0.92, and -0.88 respectively. As shown in Figure 2 (b), it can be found that the linear fitting results under each incidence angle have little difference as a whole, and are not dependent on the incidence angle.

[0061] Step 5: The altimeter and buoy effective wave height data are used to evaluate and verify the synthetic aperture radar effective wave height estimation value, the average deviation of the synthetic aperture radar effective wave height estimation value at each incidence angle compared with the altimeter effective wave height is compared, and the deviation and standard deviation of the synthetic aperture radar effective wave height estimation value compared with the buoy effective wave height are calculated.

[0062] The operation process for the result evaluation and verification is as follows:

[0063] Step 101: The altimeter effective wave height is used to evaluate the synthetic aperture radar effective wave height estimation value, and the average deviation of the synthetic aperture radar effective wave height estimation value at each incidence angle compared with the altimeter effective wave height, also known as the weight deviation, is compared. The calculation formula is as follows:

[0064]

[0065] Wherein, w1, w2,..., wn are the weight deviations of the synthetic aperture radar effective wave height estimation value at each incidence angle compared with the altimeter effective wave height, and n is the number of incidence angles. nare the weights of the synthetic aperture radar effective wave height estimation value, x1, x2, ... ..., x n is the median value of the synthetic aperture radar effective wave height estimate for each 0.2m interval, is the weight bias of the synthetic aperture radar significant wave height estimate.

[0066] Step 102: Use the on-site observed significant wave height data to verify the estimated SAR significant wave height, and calculate the deviation and standard deviation of the estimated SAR significant wave height compared to the on-site observed significant wave height.

[0067] The deviation is calculated as follows:

[0068] bias(x)=yf(x)

[0069] Where y is the estimated effective wave height of synthetic aperture radar, f(x) is the effective wave height observed in situ, and bias(x) is the deviation of the estimated effective wave height of SAR compared with the effective wave height observed in situ.

[0070] The formula for calculating the standard deviation is as follows:

[0071]

[0072] Where X is a single SAR significant wave height estimate, μ is the mean of the SAR significant wave height estimates, N is the total number of samples of the SAR significant wave height estimates, and STD is the standard deviation of the SAR significant wave height estimates.

[0073] In one embodiment, the buoy significant wave height uses the significant wave height data of the global coastal waters from 2019 to 2022 from the National Data Buoy Center of the National Oceanic and Atmospheric Administration of the United States.

[0074] Compared with the altimeter significant wave height, the SAR significant wave height estimate has a bias within the range [-2, 2] at all incidence angles, with most of the bias concentrated between [-0.5, 0.5]. The calculated SAR significant wave height weighted biases for the corresponding incidence angle ranges are 0.07982, -0.05852, -0.13482, -0.0728, and -0.14888. Compared with buoy significant wave height, using approximately 4,000 significant wave height data from the National Data Buoy Center of the National Oceanic and Atmospheric Administration (NOAA), the SAR significant wave height estimate has a bias of -0.29 m and a standard deviation of 1.28 m. Most of the bias is concentrated between [-0.6, 0]. Evaluation and verification demonstrate that the SAR significant wave height estimate has very little error compared to the altimeter and buoy significant wave heights, demonstrating its accuracy and reliability.

Claims

1. A method for extracting effective wave height based on synthetic aperture radar images, characterized in that: The following steps are involved: Step 1: Obtain the SAR interferometry imaging mode and altimeter significant wave height data for the sea area studied for many years, match the altimeter significant wave height with the SAR interferometry imaging mode data, set an appropriate time matching window and spatial matching distance, and obtain a matched sample data set; Step 2: Select the altimeter significant wave height data of the sample dataset, analyze the global distribution of the altimeter significant wave height, divide it into different incidence angle ranges, count the data volume of the altimeter significant wave height in different incidence angle ranges, and obtain the numerical distribution of the altimeter significant wave height in the global near-coast range; Step 3: Perform quality control on the normalized variance of the synthetic aperture radar backscatter coefficient and the altimeter significant wave height of the sample data set. Use the confidence ellipse fitting method, set quality control conditions, and perform confidence ellipse fitting to obtain the fitted significant wave height and normalized variance data. Step 4: Perform linear fitting on the fitted significant wave height and normalized variance data at different incident angles to obtain the fitting coefficients at each incident angle. A linear relationship between significant wave height and normalized variance is established, and the normalized variance is input to obtain the estimated value of the synthetic aperture radar significant wave height through the model. Step 5: Use the altimeter and buoy significant wave height data to evaluate and validate the SAR significant wave height estimate. Compare the average deviation of the SAR significant wave height estimate compared to the altimeter significant wave height at each incidence angle. Calculate the deviation and standard deviation of the SAR significant wave height estimate compared to the buoy significant wave height.

2. The method for extracting significant wave height based on synthetic aperture radar images according to claim 1, wherein: In step 1, the altimeters used include Cryosat-2 and Sentinel-3_a launched by the European Space Agency, Jason-2 launched by the Delta-2 carrier rocket, Jason-3 launched by the Falcon 9 carrier rocket, and SARAL launched by the French National Center for Space Studies and the Indian Space Research Institute. Their resolution is 6 km, and significant wave height data near the coast of the world from 2019 to 2022 are selected.

3. The method for extracting significant wave height based on synthetic aperture radar images according to claim 2, wherein: In step 1, the synthetic aperture radar used is the interferometric imaging mode of the Sentinel-1 in the European Space Agency's Copernicus program, with a resolution of 10m and a synthetic aperture radar image size of 250kmx250km.

4. The method for extracting significant wave height based on synthetic aperture radar images according to claim 3, wherein: The time matching window and spatial matching distance are as follows: in terms of time, ±1.5 hours is used as the time matching window; in terms of space, the data points on the synthetic aperture radar image are matched within a circular range with a radius of 3 km and the center of the circle obtained by the altimeter as the center of the circle with a radius of 3 km. The altimeter circular range with a synthetic aperture radar image pixel less than 500 is eliminated, and the matching is completed.

5. The method for extracting effective wave height based on synthetic aperture radar image according to claim 1, characterized in that ; In step 2, the sample significant wave height data are distributed in the near-coast areas of the world, with the incident angle ranges of 32°±1°, 34°±1°, 36°±1°, 38°±1° and 40°±1°. The sample data sets exceed 67,000, of which more than 12,000 are within the incident angle range of 38°±1°, and 6,000 ±100 are within the incident angle range of 32°±1°. In the near-coast area of ​​the world, the significant wave height is mainly concentrated around 2m±0.5, with more than 5,000 data points, showing a Weibull distribution.

6. The method for extracting significant wave height based on synthetic aperture radar images according to claim 5, characterized in that: The global near coasts include the coast of Hawaii, the east and west coasts of the United States, the west coast of Europe, the southeast coast of Africa, the southeast coast of China and the coast of Japan.

7. The method for extracting significant wave height based on synthetic aperture radar images according to claim 1, wherein: In step 3, the quality control is performed on all data points using a confidence ellipse fitting method, where the normalized variance consists of the mean and the standard deviation, and the formula is as follows: Where σ is the standard deviation, μ is the mean, and Nvar is the normalized variance.

8. The method for extracting significant wave height based on synthetic aperture radar images according to claim 1, wherein: In step 4, the linear fitting is a one-variable first-order polynomial fitting, and the calculation formula is: Where y is the estimated value of the effective wave height of the synthetic aperture radar, x is the normalized variance, and are the fitted coefficients of the polynomial.

9. The method for extracting significant wave height based on synthetic aperture radar images according to claim 1, wherein: In step 5, the buoy significant wave height uses the significant wave height data of global nearshore waters from 2019 to 2022 from the National Data Buoy Center of the National Oceanic and Atmospheric Administration of the United States.

10. The method for extracting significant wave height based on synthetic aperture radar images according to claim 9, characterized in that: In step 5, the specific method includes the following steps: Step 101: Use the altimeter significant wave height to evaluate the estimated SAR significant wave height. Compare the average deviation of the SAR significant wave height estimate at each incident angle compared to the altimeter significant wave height, also known as the weighted deviation. The calculation formula is: in, is the weight of the synthetic aperture radar effective wave height estimate, is the median value of the synthetic aperture radar effective wave height estimate for each 0.2m interval, is the weight bias of the synthetic aperture radar significant wave height estimate; Step 102: Verify the estimated SAR significant wave height using the field observed significant wave height data, and calculate the deviation and standard deviation of the estimated SAR significant wave height compared to the field observed significant wave height. The deviation is calculated as follows: in, is the estimated value of synthetic aperture radar effective wave height, For on-site observation of significant wave height, is the deviation of the estimated effective wave height of synthetic aperture radar compared with the effective wave height of field observation, The formula for calculating the standard deviation is as follows: in, is the estimated value of the effective wave height of a single synthetic aperture radar, is the average of the synthetic aperture radar significant wave height estimates, is the total number of samples of SAR effective wave height estimation, and STD is the standard deviation of SAR effective wave height estimation.

Images