Airborne SAR distance compressed domain sea clutter suppression method based on generalized gamma distribution

Abstract

The invention particularly relates to an airborne SAR distance compressed domain sea clutter suppression method based on generalized gamma distribution, and the method comprises the steps: carrying out the range pulse compression processing of an echo signal received by an airborne SAR, and extracting a pure sea clutter sample from compressed data through employing a non-local sampling strategy; then, a generalized gamma distribution model is quickly fitted for the clutter samples by using an analytical method based on a second type logarithmic cumulant, and a self-adaptive detection threshold is solved through an incomplete gamma function based on the model and a set false alarm probability. And finally, according to the threshold value, carrying out judgment and filtering on distance compressed domain data, suppressing clutter regions lower than the threshold value, and meanwhile, retaining potential target signals, thereby realizing efficient and steady clutter suppression under complex sea conditions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of synthetic aperture radar (SAR) signal processing, in particular to a method for sea clutter suppression in the range compression domain of airborne SAR based on generalized gamma distribution. It is suitable for data processing and target detection of airborne SAR system in marine monitoring, and can be widely applied to marine target detection and marine traffic monitoring scenes. BACKGROUND

[0002] With the rapid development of radar imaging and target recognition technology, airborne synthetic aperture radar (SAR) has become an important sensing means in key fields such as marine monitoring, marine law enforcement and disaster assessment, due to its all-weather, all-day and high-resolution imaging capabilities. Airborne SAR can realize electromagnetic detection of a large area of sea through a high-speed flight platform, and can continuously observe ships, icebergs, floating objects and sea surface conditions under complex weather and lighting conditions. However, in the signal receiving process of the airborne SAR system, there are a large number of randomly fluctuating clutter signals in the sea surface echo, and the amplitude distribution usually shows heavy-tailed, non-Gaussian and non-stationary characteristics, and changes significantly with the incident angle, sea state and wind wave state. Such clutter often masks the echo of small ship targets, resulting in decreased target detection performance and increased false alarm rate, reducing the usability of SAR in complex sea conditions. Therefore, researching a method that can accurately represent the statistical characteristics of sea clutter and achieve robust suppression is of great significance to improve the target detection capability and system reliability of airborne SAR in complex sea conditions.

[0003] Currently, the research on sea clutter suppression mainly focuses on two directions: clutter statistical modeling and adaptive detection. The traditional constant false alarm rate detection method estimates the background power through a sliding window and sets a threshold, which has the advantages of simple implementation and good real-time performance, but it is prone to detection failure in non-uniform background or the presence of strong scattering targets. To more accurately describe the statistical characteristics of sea clutter, researchers have proposed a variety of probability models, such as Rayleigh distribution, Weibull distribution, K distribution, lognormal distribution and generalized gamma distribution (GΓD). Among them, the GΓD model has a flexible structure with three adjustable parameters, and can unify the description of various distribution forms, becoming a research hotspot for sea clutter modeling in recent years. However, existing methods generally have problems such as complex calculation, insufficient real-time performance, unstable parameter estimation, and are difficult to balance model accuracy and engineering application efficiency. Therefore, there is an urgent need for a suppression method that can quickly complete parameter estimation and threshold calculation under limited computing resources, and is suitable for airborne SAR sea clutter environment.

[0004] The prior art proposes a sea clutter suppression method combining wavelet neural network and intelligent optimization algorithm. Firstly, the method uses the nonlinear modeling and self-learning ability of the wavelet neural network to model and predict the measured sea clutter data, thereby realizing effective estimation and suppression of the clutter. Then, in the network training process, the particle swarm optimization algorithm and the gravitational search algorithm are introduced into the joint optimization strategy, the global search ability of the particle swarm optimization algorithm and the local convergence ability of the gravitational search algorithm are fused, and the weights and parameters of the wavelet neural network are optimized to improve the prediction accuracy and convergence speed of the model. Finally, the optimized wavelet neural network model is applied to the prediction of sea clutter data, the prediction residual is analyzed, and the detection and suppression of weak target signals in the clutter background are realized. The advantage of this method is that it combines the nonlinear fitting ability of the neural network and the global optimization characteristics of the intelligent optimization algorithm, and can better adapt to the complex sea clutter environment. However, the disadvantage is that the network training and optimization process has a huge amount of calculation, and the parameter setting is complex, which is difficult to meet the stringent requirements of real-time processing of airborne SAR systems.

[0005] The prior art also proposes a sea clutter suppression method. Firstly, the method acquires the sea surface range-time image according to the non-uniform and non-Gaussian characteristics of the sea surface echo, and calculates the autocorrelation function of each range cell along the time dimension to construct the autocorrelation matrix; then, the covariance matrix of the speckle component is constructed according to the autocorrelation matrix, and the characteristic vector of the speckle component is extracted by eigenvalue decomposition, thereby forming the characteristic subspace of the sea clutter; finally, the sea clutter subspace is substituted into the orthogonal projection suppression algorithm, and the sea clutter component in the original image is suppressed by the orthogonal projection operator, thereby realizing the enhancement of the target signal. The advantage of this method is that it can directly use the received data to construct the sea clutter characteristic subspace, without relying on the adjacent reference cell, avoiding the inaccuracy of covariance matrix estimation, and having good adaptability and engineering feasibility. However, its limitation is that the method relies heavily on autocorrelation calculation and matrix decomposition, and in the case of large-scale data or high-resolution SAR, matrix decomposition and other operations will bring heavy computational burden, which restricts its application in real-time processing systems.

[0006] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present application, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0007] The present application aims at the deficiencies of the prior art, and provides a sea clutter suppression method in a SAR range compression domain based on a generalized gamma distribution, so as to solve the problems of poor adaptability of a traditional clutter model in a non-imaging data domain, high parameter estimation complexity, and unstable suppression effect under strong sea state conditions, thereby realizing effective modeling and suppression of sea clutter in SAR range compression data, improving the real-time performance and robustness of signal processing, and providing a reliable data basis for subsequent target detection and ocean monitoring.

[0008] Other characteristics and advantages of the present application will become apparent from the following detailed description, or will be learned by practice of the present application.

[0009] According to a first aspect of the present application, a sea clutter suppression method in a SAR range compression domain based on a generalized gamma distribution is provided, and the method comprises: receiving a SAR echo signal, performing range direction pulse compression processing on the received SAR echo signal to obtain range compression domain data; performing block processing on the range compression domain data, and constructing a representative sea clutter sample set by using a non-local sample extraction strategy; statistically modeling the range compression domain sea clutter samples, and establishing a generalized gamma distribution probability model conforming to the distribution characteristics thereof; calculating the log-cumulant of the sea clutter samples, and extracting the steady-state statistical characteristics thereof; calculating the parameters of the generalized gamma distribution model in the range compression domain based on the second type of cumulant, so as to obtain the shape parameter, the scale parameter and the power parameter of the generalized gamma distribution; under the condition of setting a false alarm rate, calculating an adaptive detection threshold based on an inverse solution of an incomplete gamma function; judging the range compression domain signal according to the adaptive detection threshold, suppressing the clutter area lower than the threshold, and keeping the target area higher than the threshold, so as to realize clutter suppression output.

[0010] In some example embodiments, the range direction pulse compression processing is specifically a convolution operation of the SAR received signal and a range direction reference function, and the range direction reference function is a negative conjugate function of the airborne SAR transmitted pulse signal.

[0011] In some example embodiments, the block processing of the range compression domain data and the construction of the representative sea clutter sample set by using the non-local sample extraction strategy are specifically as follows: the range compression data is divided along the azimuth direction, and the interval is twice the synthetic aperture length; the range compression data is divided along the range direction, and the interval is twice the radar pulse length; The current data block to be processed is taken as a target region, and several azimuth units are selected at fixed intervals in the target region; The maximum value in each azimuth unit is removed; The remaining data in the target region is taken as a sea clutter sample set.

[0012] In some example embodiments, the logarithmic cumulants of the sea clutter samples are calculated, specifically: By taking the natural logarithm of the amplitude data, a sea clutter statistical sequence in the logarithmic domain is obtained, and the first three order empirical logarithmic cumulants are calculated by statistically processing the sample sequence in the logarithmic domain The specific calculation method of the estimate value is as follows:

[0013]

[0014]

[0015] wherein, is the sea clutter data obtained by sampling, represents the number of data samples, and the variable above represents the estimate value of the second type of cumulant.

[0016] In some example embodiments, the distance compression domain generalized gamma distribution model parameter calculation is realized based on the second type of cumulant, and the shape parameter, scale parameter and power parameter of the generalized gamma distribution are obtained, specifically: The shape parameter is calculated as

[0017]

[0018]

[0019]

[0020] The power parameter is calculated as

[0021] After the estimate value of is obtained, the estimate value of is calculated as

[0022] wherein, is the first order Polygamma function.

[0023] The scale parameter is calculated as ​

[0024] After obtaining and , the estimated value of the scale parameter is calculated by the following formula

[0025] wherein, is the logarithmic derivative of the gamma function.

[0026] In some example embodiments, the adaptive detection threshold is calculated by the following formula:

[0027] wherein, is usually a constant less than 0.01, represents the inverse function of the incomplete gamma function, wherein, is the Doppler bandwidth, is the transmission time of the transmitted wave to the ground.

[0028] In some example embodiments, the method further comprises constructing a binary matrix for the decision result, and performing sliding window filtering processing on the binary matrix to eliminate isolated highlights and random noise.

[0029] According to a second aspect of the present application, there is provided a storage medium having stored thereon a computer program, which, when executed by a processor, implements the airborne SAR sea clutter suppression method in range-compression domain based on generalized gamma distribution according to the first aspect.

[0030] According to a third aspect of the present application, there is provided a computer program product having stored thereon a computer program, which, when executed by a processor, implements the airborne SAR sea clutter suppression method in range-compression domain based on generalized gamma distribution according to the first aspect.

[0031] According to a fourth aspect of the present application, there is provided an electronic device comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to implement the airborne SAR sea clutter suppression method in range-compression domain based on generalized gamma distribution according to the first aspect by executing the executable instructions.

[0032] The airborne SAR sea clutter suppression method in range-compression domain based on generalized gamma distribution provided by the embodiments of the present application has the following beneficial effects compared with the prior art: ​First, the parameter estimation is accurate and efficient. The application introduces an analytical estimation method based on the second type of logarithmic cumulant in the modeling of the generalized gamma distribution, which avoids the traditional numerical iterative process and can directly solve the model parameters in a closed form. This method significantly reduces the computational complexity while ensuring the accuracy of the estimation, which is suitable for real-time processing requirements of airborne platforms and solves the problems of existing algorithms such as dependence on initial value, slow convergence, and heavy computational burden.

[0033] Second, the adaptive and robust suppression mechanism. The application establishes an adaptive threshold calculation model based on the incomplete gamma function, which can automatically adjust the detection threshold according to different sea conditions and radar operating parameters, and realize the dynamic suppression of sea clutter energy in complex background. This mechanism can effectively weaken the non-uniform interference such as sea wave spikes and isolated strong scattering points, improve the robustness and detection reliability of clutter suppression, and provide stable data input for subsequent target detection and imaging processing.

[0034] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0035] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained from these drawings without creative labor for those skilled in the art.

[0036] Figure 1 Flowchart of the method of the present application; Figure 2 Effect comparison chart of each amplitude clutter fitting result of the present application; Table 1 is a quantitative comparison table of clutter fitting results of the present application. DETAILED DESCRIPTION

[0037] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the gist of the example implementations to those skilled in the art. The features, structures, or characteristics described in connection with the examples can be combined in any suitable manner in one or more implementations.

[0038] Furthermore, the accompanying drawings are included to provide a further understanding of the application, and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments of the application and, together with the description, serve to explain the principles of the application. In the drawings:

[0039] The idea of the present application is that firstly, the echo signal received by airborne SAR is processed in range direction pulse compression, and a non-local sampling strategy is used to extract pure sea clutter samples from the compressed data. Then, an analytical method based on the second type of logarithmic cumulant is used to quickly fit the generalized gamma distribution model for these clutter samples, and based on the model and the set false alarm probability, the adaptive detection threshold is solved by the incomplete gamma function. Finally, according to the threshold, the range compression domain data is judged and filtered, and the clutter area below the threshold is suppressed, while the potential target signal is retained, so as to realize efficient and robust clutter suppression in complex sea conditions.

[0040] The main innovation and advancement of the present application are embodied in the following two aspects: 1. Cross-domain modeling and fast parameter estimation: The present application first introduces the generalized gamma distribution model into the SAR range compression domain, proposes a sea clutter modeling method of the generalized gamma distribution model in the range compression domain, and realizes fast estimation of model parameters combined with the cumulant analytical algorithm. Compared with the traditional clutter modeling method which relies on numerical iteration, this scheme significantly reduces the computational complexity and improves the real-time performance and stability.

[0041] 2. Adaptive threshold suppression mechanism: Based on the parameterized modeling, the present application proposes an adaptive threshold calculation method based on the incomplete gamma function, which can automatically adjust the suppression threshold under different sea conditions and sensor conditions, effectively weaken the transient strong interference such as sea wave peak, and realize more robust sea clutter suppression effect.

[0042] The sea clutter suppression method in the range compression domain of airborne SAR based on the generalized gamma distribution proposed by the present application firstly processes the echo signal received by airborne SAR in range direction pulse compression, and uses a non-local sampling strategy to extract pure sea clutter samples from the compressed data. Then, an analytical method based on the second type of logarithmic cumulant is used to quickly fit the generalized gamma distribution model for these clutter samples, and based on the model and the set false alarm probability, the adaptive detection threshold is solved by the incomplete gamma function. Finally, according to the threshold, the range compression domain data is judged and filtered, and the clutter area below the threshold is suppressed, while the potential target signal is retained, so as to realize efficient and robust clutter suppression in complex sea conditions.

[0043] refer to Figure 1 As shown, the specific steps may include: Step 1: Receive the raw echo signal from the airborne SAR and complete pulse transmission and echo acquisition; Step 2: Perform range pulse compression processing on the received SAR echo signal to obtain range compressed domain complex amplitude data; Step 3: Divide the range compression domain data into blocks and construct a representative sea clutter sample set using a nonlocal sample extraction strategy; Step 4: Perform statistical modeling on the sea clutter samples in the range-compressed domain to establish a probabilistic model framework that conforms to their distribution characteristics; Step 5: Calculate the log-cumulative value of the sea clutter sample and extract its steady-state statistical characteristics; Step 6: Calculate the parameters of the generalized gamma distribution model in the distance compression domain based on the second type of cumulants, and obtain the shape parameter, scale parameter and power parameter of the generalized gamma distribution; Step 7: Under the condition of setting the false alarm rate, calculate the adaptive detection threshold based on the inverse solution of the incomplete gamma function; Step 8: Based on the threshold, the distance compression domain signal is judged, the clutter area below the threshold is suppressed, and the target area above the threshold is maintained, so as to achieve clutter suppression output.

[0044] This invention innovatively establishes a generalized gamma distribution statistical model in the range compression domain, and combines logarithmic cumulant parameter estimation with gamma function threshold inverse solution to achieve integrated processing of sea clutter modeling, parameter solving, and threshold suppression. Furthermore, this method avoids the multiple iterations required for traditional parameter solving, resulting in low computational cost, stable convergence, and strong real-time performance. It can effectively reduce sea clutter interference and improve airborne SAR target detection performance and imaging quality under complex sea conditions.

[0045] The steps in this exemplary embodiment will now be described in more detail with reference to the accompanying drawings and embodiments.

[0046] Step 1: Airborne SAR echo signal reception.

[0047] During operation, the airborne SAR system receives ground echo signals, and its transmitted signal model is represented as follows:

[0048] In the formula, For the sake of speed, For the transmit pulse width, Let be the range envelope function of the signal. For carrier frequency, It is the linear frequency modulation frequency.

[0049] Generally, the single pulse echo of airborne SAR radar is expressed in the baseband signal at the receiving end as follows:

[0050] In the formula, is the backscattering coefficient of the scanning target, is a distance envelope function, is an azimuth envelope function, is the speed of light, is the distance between the target and the radar.

[0051] Step 2: Distance direction pulse compression processing of the airborne SAR echo signal.

[0052] This step performs distance direction pulse compression operation on the received SAR original echo signal. The essence of this process is a matched filtering operation, that is, the SAR received signal is correlated with the distance direction reference function to realize the focusing of the target echo energy in the distance direction.

[0053] In the present application, the distance direction reference function is the negative conjugate function of the airborne SAR transmitted pulse signal, and its expression is as follows:

[0054] In the formula, the symbol represents a complex conjugate operation.

[0055] The received echo signal is convolved with the reference function to obtain the distance compressed airborne SAR echo signal as follows:

[0056] In the formula, is the distance direction echo signal bandwidth, represents a convolution operation.

[0057] After the processing of this step, the scattering energy of the target signal is focused in the distance direction to its nearest slant range position , and this position is independent of the azimuth direction frequency, thereby realizing the accurate compression of the distance direction signal and providing input data for subsequent sea clutter sample extraction and statistical modeling.

[0058] Step 3: Distance compression domain data block and non-local sea clutter sample extraction.

[0059] This step aims to extract representative sea clutter samples from range-compressed SAR data, providing input for subsequent statistical modeling based on the generalized gamma distribution. Considering the non-stationarity of sea clutter and its statistical differences in azimuth and range directions, this invention employs a sample construction method based on nonlocal features. By dividing the range-compressed data into blocks and performing sparse sampling, reliable sample acquisition is achieved.

[0060] SAR range compressed domain data In a two-dimensional plane The above exhibits the following characteristics: signal energy is concentrated and clearly focused along the range direction; energy distribution is sparse and random along the azimuth direction. Therefore, this invention proposes a nonlocal sampling method that is dense along the range direction and sparse along the azimuth direction, the specific steps of which are as follows: 1) Divide the range compression data along the azimuth direction at intervals of twice the synthetic aperture length; 2) Divide the range compression data along the range direction at intervals of twice the radar pulse length; 3) Take the current data block to be processed as the target region, and sparsely select data within it at fixed intervals (e.g., every M distance units). One directional unit; 4) Remove the maximum value in each azimuth cell; 5) Use the remaining data in the target area as a clutter sample set.

[0061]

[0062] This step yields a sea clutter sample set for modeling. This set contains only background clutter amplitude information, which can effectively reflect the statistical characteristics of sea clutter in the range-compressed domain.

[0063] Step 4: Model sea clutter in the range compressed SAR data domain.

[0064] This step aims to establish a statistical model of sea clutter applicable to the range compression domain of airborne SAR based on the generalized gamma distribution theory, referred to as RC-GΓD, to describe the probability distribution characteristics of echo amplitude under different sea states.

[0065] The generalized gamma distribution is a three-parameter statistical model that uses a scaling parameter. Power parameters With shape parameters The probability density function is defined as follows to flexibly describe the distribution characteristics of random signals:

[0066] In the formula, Gamma function. Theoretical analysis and experiments show that the statistical distribution of clutter amplitude in the range-compressed domain still obeys the generalized gamma distribution family, but the scale parameter will change. Therefore, the generalized gamma distribution model in the image domain can be mapped to the range-compressed domain through a scaling coefficient, and the generalized gamma distribution model in the range-compressed domain is obtained as follows:

[0067] wherein, After modeling in this step, the RC-GΓD model obtained can accurately describe the statistical law of sea clutter in the range-compressed domain.

[0068] Step 5: Calculate the log-cumulants of the sea clutter sample.

[0069] The purpose of this step is to use the sea clutter sample data in the range-compressed domain to extract stable quantitative indicators reflecting the statistical characteristics thereof, thereby providing a basis for the inversion of the model parameters in the subsequent step.

[0070] First, let be the sea clutter data obtained by sampling, wherein represents the number of data samples. For this sample set, the statistical sequence of the clutter in the logarithmic domain is obtained by taking the natural logarithm of the amplitude data, so as to weaken the influence of outliers and enhance the stability of parameter estimation.

[0071] Subsequently, the sample sequence is statistically processed in the logarithmic domain to calculate the first three order empirical log-cumulants The specific calculation method of the estimated value is as follows:

[0072]

[0073]

[0074] wherein, the above the variable represents the estimated value of the second type of cumulant. This calculation process does not require integral or derivative operations, but only the mean value and central moments of the sample logarithm.

[0075] Through this step, the empirical cumulants of the RC-GΓD model can be obtained. These statistical quantities can effectively reflect the statistical characteristics of the sea clutter in the range-compressed domain, and serve as the input quantities for solving the model parameters in the next step, for the inversion of the shape parameter, the power parameter and the scale parameter of the generalized gamma distribution.

[0076] Step 6: RC-GΓD model parameter calculation based on the second type of cumulant.

[0077] The purpose of this step is to solve for the three key parameters of the distance-compressed domain generalized gamma distribution model based on the second type of log-cumulant calculated in the previous step: shape parameter. Power parameters and scale parameters These parameters determine the shape of the sea clutter probability distribution, its decay rate, and its overall energy level.

[0078] Step 6a: Calculate shape parameters

[0079] To obtain a closed-form solution for the shape parameters with respect to the sampled values, it is first necessary to solve for the coefficients.

[0080] Then, calculate the intermediate variables.

[0081] Finally, the shape parameters can be calculated. Closed-form solution

[0082] At this point, the estimated value calculated in step 5 is... and You can get the result directly. The estimated value .

[0083] Step 6b: Calculate the power parameter

[0084] In obtaining The estimated value back, The estimated value It can be calculated using the following formula:

[0085] Step 6c: Calculate scale parameters

[0086] In obtaining and Then, scale parameters The estimated value It can be calculated using the following formula:

[0087] This step yields the complete parameter set for the RC-GΓD model. , and This provides the basic input for the subsequent step 7, adaptive threshold calculation and sea clutter suppression based on RC-GΓD.

[0088] Step 7: Adaptive threshold calculation based on incomplete Gamma function.

[0089] This step aims to calculate the detection threshold in range-compression domain using the established RC-GΓD sea clutter model. Based on the analytical property of incomplete Gamma function, the invention adopts a method to calculate the threshold without iteration.

[0090] Firstly, based on the estimation results of the aforementioned parameters , and , the cumulative distribution function of sea clutter in range-compression domain can be expressed as

[0091] where is the incomplete Gamma function, which is defined as

[0092] In the above formula, is the formal parameter of the function, which is used to explain the calculation method of the function, is a virtual variable used in integral calculation, which has no specific physical meaning itself and is only used temporarily in the integral process.

[0093] Then, according to the definition of false alarm probability , the adaptive threshold can be calculated by the following analytical expression

[0094]

[0095] where is usually set as a constant less than 0.01, represents the inverse function of the incomplete Gamma function, and the calculation method of can be calculated according to the asymptotic expression method:

[0096] This approximation converges rapidly at low false alarm rate and has low computational complexity.

[0097] Step 8: Sea clutter suppression based on adaptive threshold.

[0098] The purpose of this step is to use the adaptive detection threshold calculated in the previous stage to suppress the sea clutter in the range-compression domain data of airborne SAR, thereby effectively distinguishing the target echo from the sea clutter component. ​

[0099] First, the distance compression domain amplitude data matrix Each pixel amplitude value According to the detection threshold Decision:

[0100] Among them, Indicates that the pixel amplitude exceeds the threshold, which is a high scattering echo point; Indicates that it is below the threshold, which is a sea clutter background. The processing result constitutes a binary matrix .

[0101] Then, for further elimination of isolated highlights and random noise, the matrix Sliding window filtering processing is performed. The local energy density is calculated in each window:

[0102] When , where is the energy threshold, which is usually set to a small value (such as 0.1) according to the false alarm probability requirement and simulation experiment, all pixels in the window are reset to zero:

[0103] After this processing, isolated noise and random peak clutter are effectively suppressed, and only echo regions with local energy continuity are retained. The data after clutter suppression is mainly used for distance compression domain SAR target detection, and can also be directly used in subsequent SAR imaging, moving target detection or focusing processing modules to realize robust target extraction of airborne SAR systems in complex sea surface scenarios.

[0104] Experimental simulation The effect of the present application will be further described below in combination with a simulation experiment.

[0105] 1. Simulation experiment conditions This experiment was performed on a computer configured with an Intel Core i7-12700K processor, 32 GB DDR5 RAM, and 1 TB NVMe SSD, with Windows 11 64-bit as the operating system. The experimental platform uses a mixed environment of MATLAB R2023b and Python 3.10 to realize signal modeling, parameter estimation, and visual analysis. The experiment relies on numerical calculation libraries such as NumPy, SciPy, and Matplotlib, and calls custom signal processing scripts in the Matlab environment to complete the distance compression and parameter fitting process. The above environment configuration ensures the processing efficiency and accuracy of the simulation data.​

[0106] 2. Simulation experiment content and result analysis: The main goal of this experiment is to verify the effectiveness and stability of the airborne SAR sea clutter suppression method in the range compression domain based on the generalized gamma distribution proposed in the present application. The experimental data is generated by a typical airborne SAR sea surface simulation model, with a working frequency of 9.6 GHz (X-band), a pulse repetition frequency (PRF) of 1500 Hz, a pulse width of 20 microseconds, a signal bandwidth of 30 MHz, and a sampling rate of 60 MHz. At the same time, sea clutter components conforming to the generalized gamma distribution characteristics are added to the simulation signal to simulate the complex background environment under actual sea conditions.

[0107] In the experimental results Figure 2 , the goodness of fit of each comparison model under two simulation scenarios is shown, where Figure 2 (a) and Figure 2 (b) are sea scenes with relatively stable sea conditions, low wind speed, and few wave peaks, Figure 2 (c) and Figure 2 (d) are sea scenes with relatively severe sea conditions. Among them Figure 2 (b) and Figure 2 (d) are probability density function curve graphs in semi-log form, Figure 2 (a) and Figure 2 (c) are complementary cumulative distribution function curve graphs. From Figure 2 (b) and Figure 2 (d), it can be seen that the Log-normal, Weibull and K+Noise distributions have poor matching with the data. Although K+Rayleigh, GΓD and 3MD have better fitting effects, they have errors in the tail fitting, which are respectively underestimation and overestimation. In contrast, the estimation results of the method proposed in the present application are highly consistent with the data. From Figure 2 (a) and Figure 2 (c), it can be seen that the threshold error of the method proposed in the present application is the smallest, dB.

[0108] In Table 1 of the experimental results, the numerical evaluation mean values of different distribution functions on each sub-scene are shown. As can be seen from the table, the method of the present application has the smallest numerical results in all evaluation indicators, showing better stability and accuracy than other distribution models. Specifically, the average KL value under scene 1 is only 0.0131, and under scene 2 is 0.0183, which is about 70% lower than the traditional Weibull distribution and K+Rayleigh model; the TKL index and the average threshold error also remain at a low level, indicating that the present application can achieve high-precision modeling and robust suppression of clutter distribution under multiple sea conditions. In addition, compared with the traditional GΓD and 3MD methods, the algorithm of the present application can obtain convergent parameters without relying on multiple iterations, significantly reducing the computational complexity, and achieving better clutter energy suppression while maintaining the integrity of the target signal. Thus, it is verified that the present application has higher adaptability and engineering practical value in complex sea clutter environment.

[0109] Table 1

[0110] It should be noted that, as another aspect, the present application also provides a storage medium, which can be included in an electronic device or exist independently without being assembled into the electronic device. The storage medium carries one or more programs, which, when executed by an electronic device, enable the electronic device to implement the method described in the following embodiments.

[0111] In one embodiment, the present application provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.

[0112] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not for limiting purposes. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be executed synchronously or asynchronously, for example, in multiple modules.

[0113] Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. The present application is intended to cover any variations, uses or adaptations of the application following, in general, the principles of the application and including such steps that are apparent to those skilled in the art to which the application pertains. The specification and examples are to be regarded as exemplary only, and the true scope and spirit of the application is indicated by the claims.

[0114] It should be understood that the application is not limited to the precise construction which has been described above and which shown in the drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application should be determined by the claims appended hereto.

Claims

1. A method for sea clutter suppression in airborne SAR range-compression domain based on generalized gamma distribution, characterized in that, The method comprises: receiving an airborne SAR echo signal, performing range direction pulse compression processing on the received SAR echo signal to obtain range compression domain data; performing block processing on the range compression domain data, and constructing a representative sea clutter sample set using a non-local sample extraction strategy; statistically modeling the range compression domain sea clutter samples to establish a generalized gamma distribution probability model that conforms to the distribution characteristics thereof; calculating the log-likelihood of the sea clutter samples to extract steady-state statistical characteristics thereof; calculating the parameters of the range compression domain generalized gamma distribution model based on the second type of cumulant to obtain the shape parameter, scale parameter and power parameter of the generalized gamma distribution; under the condition of a set false alarm rate, calculating an adaptive detection threshold based on an inverse incomplete gamma function; judging the range compression domain signal based on the adaptive detection threshold, suppressing the clutter region below the threshold, and keeping the target region above the threshold to achieve clutter suppression output.

2. The method of claim 1, wherein, The range direction pulse compression processing specifically comprises performing convolution operation on the SAR receiving signal and a range direction reference function, wherein the range direction reference function is the negative conjugate function of the airborne SAR transmitting pulse signal.

3. The method of claim 1, wherein, The block processing on the range compression domain data and the construction of the representative sea clutter sample set using the non-local sample extraction strategy specifically comprise: dividing the range compression data along the azimuth direction with an interval of twice the synthetic aperture length; dividing the range compression data along the range direction with an interval of twice the radar pulse length; taking the current data block to be processed as a target region, and sparsely selecting a plurality of azimuth units at fixed intervals in the target region; removing the maximum value in each azimuth unit; taking the remaining data in the target region as the sea clutter sample set.

4. The method of claim 1, wherein, The calculation of the log-likelihood of the sea clutter samples specifically comprises: The clutter statistical sequence in logarithm domain is obtained by taking natural logarithm of the amplitude data, and the sample sequence is statistically processed in the logarithm domain to calculate the first three order empirical logarithm cumulants The specific calculation method of the estimation value is as follows: wherein is the sampled sea clutter data, denotes the number of data samples, the variable above denotes the estimate of the second kind of cumulant.

5. The method of claim 4, wherein, The calculation of the parameters of the range compression domain generalized gamma distribution model based on the second type of cumulant to obtain the shape parameter, scale parameter and power parameter of the generalized gamma distribution specifically comprises: Computing shape parameters Computing the power parameter After obtaining the estimate of the estimate of the estimate of the estimate of is calculated by the equation wherein is the 1st Polygamma function. Computing scale parameters In obtaining and Then, scale parameters The estimated value Calculated using the following formula: wherein is the logarithmic derivative of the gamma function.

6. The method of claim 5, wherein, The calculation of the adaptive detection threshold uses the following formula: wherein is generally set to a constant less than 0.01, denotes the inverse function of the incomplete gamma function, wherein, is the Doppler bandwidth, is the transmission time of the transmitted wave to the ground.

7. The method of claim 1, wherein, The method further comprises constructing a binary matrix for the judgment result, and performing sliding window filtering processing on the binary matrix to eliminate isolated highlights and random noise.

8. A storage medium having stored thereon a computer program, characterized in that The computer program, when executed by a processor, implements the airborne SAR range compression domain sea clutter suppression method based on the generalized gamma distribution according to any one of claims 1 to 7.

9. A computer program product comprising a computer program, characterized in that, The computer program, when executed by a processor, implements the airborne SAR range compression domain sea clutter suppression method based on the generalized gamma distribution according to any one of claims 1 to 7.

10. An electronic device, comprising: comprise: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the airborne SAR range compression domain sea clutter suppression method based on the generalized gamma distribution according to any one of claims 1 to 7 by executing the executable instructions.

Images