Sea clutter suppression method based on self-correlation characteristics

Abstract

A sea clutter suppression method based on self-correlation features belongs to the field of image processing of measurement signals, and comprises the following steps: firstly, obtaining a self-correlation function of an RT distance-time image of a sea surface and constructing a self-correlation matrix, then constructing a covariance matrix of speckle components according to the self-correlation matrix, and performing eigenvalue decomposition on the covariance matrix of the speckle components to obtain an eigenvalue decomposition matrix; obtaining a feature vector matrix of the speckle components; then, according to the number of feature vectors of a covariance matrix of the sea surface RT distance-time image, obtaining a subspace of sea clutters; and finally, substituting the subspace of the sea clutter into an orthogonal projection suppression algorithm to obtain an orthogonal projection operator, and effectively suppressing the sea clutter in the sea surface RT distance-time image by using the orthogonal projection operator. The method can directly utilize the received data to construct the feature subspace of the sea clutter, does not need to depend on reference data of adjacent units, and avoids inaccuracy in covariance matrix estimation.

Description

Technical Field

[0001] The invention relates to the field of image processing of measurement signals, in particular to a sea clutter suppression method based on autocorrelation characteristics. Background Art

[0002] With the increasing demand for real-time monitoring and precise detection of marine targets, it is increasingly important to be able to quickly and reliably detect small maneuvering targets on the sea surface (such as boats, frogmen, low-altitude aircraft, etc.). However, affected by the movement of the sea surface, sea clutter exhibits non-uniform, non-Gaussian and non-stationary characteristics, resulting in sea surface targets being interfered with by sea clutter in both the time domain and the frequency domain. At present, the research on small sea surface target detection mainly focuses on two aspects, one is the research on sea clutter suppression methods, and the other is the research on sea surface target detection algorithms. Since sea clutter suppression is the prerequisite for effective target detection, it is more important.

[0003] At present, the methods for suppressing sea clutter include time domain cancellation, wavelet transform, deep learning, space-time adaptive processing and subspace methods. The time domain cancellation method generates a cancellation signal based on the measured sea clutter estimation parameters and subtracts the signal from the measured data to achieve sea clutter suppression. However, this method relies on accurately estimating the characteristics of sea clutter. Any deviation may weaken the characteristics of the target echo, resulting in insignificant sea clutter suppression effect. The wavelet transform rule separates sea clutter from target signals in the wavelet domain and removes clutter using threshold segmentation, but this method relies on appropriate basis functions and destroys the phase information of the received signal. With the development and application of deep learning, researchers have proposed new sea clutter suppression methods from the perspective of image processing. Although these methods perform well in some cases, they still have problems such as complex models and poor generalization, which limits their wide application. Space-time adaptive processing (STAP) can make full use of multi-domain information such as space, Doppler, and distance to suppress sea clutter. Although this method provides performance improvements to varying degrees, space-time adaptive processing is mainly applicable to airborne conditions and has a large amount of calculation. In contrast, the subspace-based sea clutter suppression algorithm has been widely used in the field of sea surface target detection due to its strong target feature extraction capability, low computational complexity and strong engineering feasibility.

[0004] Subspace methods usually decompose the echo signal into sea clutter subspace and signal subspace, set the singular values ​​in the sea clutter subspace to zero, and then reconstruct the signal to achieve sea clutter suppression. Subspace methods usually rely on adjacent reference cells to estimate the covariance matrix of sea clutter. This strategy assumes that sea clutter is uniform, that is, the backscatter in each range gate has the same statistical distribution. However, in practical applications, the training samples of adjacent range gates often do not satisfy the assumption of independent and identical distribution. Therefore, the estimation of the sea clutter covariance matrix may be inaccurate, thus affecting the effect of sea clutter suppression. Specifically, subspace suppression methods face the following problems: (1) The accuracy of covariance matrix estimation has a great impact on performance. For example, the non-uniformity of sea clutter leads to a small number of uniform sea clutter samples used for estimation, resulting in large estimation errors; (2) In actual detection, the target Doppler is unknown, and Doppler mismatch is prone to occur between the echo signal and the model, which in turn affects the detection performance; (3) The non-stationarity of the signal within the coherent processing interval (CPI) also greatly limits the performance of this type of detector. Summary of the invention

[0005] In order to solve the above problems existing in the practical application of subspace-based suppression methods, the present invention provides a sea clutter suppression method based on autocorrelation features. The method can directly use the sea surface RT range-time image to construct the covariance matrix of sea clutter without relying on the reference data of neighboring cells, thus avoiding the inaccuracy in the covariance matrix estimation.

[0006] The technical solution adopted by the present invention to solve the above technical problems is: a sea clutter suppression method based on autocorrelation characteristics, comprising the following steps: 1) Calculate the autocorrelation function of each distance unit in the sea surface RT range-time image along the time dimension, and average the results of all distance units to obtain the autocorrelation function of the sea surface RT range-time image, and then construct the autocorrelation matrix based on the autocorrelation function; 2) Construct the covariance matrix of the speckle component based on the autocorrelation matrix; 3) Perform eigenvalue decomposition on the covariance matrix of the speckle component to obtain the eigenvector matrix of the speckle component; 4) Obtain the subspace of sea clutter based on the number of eigenvectors of the sea surface RT range-time image; 5) Substitute the subspace of sea clutter into the orthogonal projection suppression algorithm to obtain the orthogonal projection operator, and use the orthogonal projection operator to effectively suppress the sea clutter in the sea surface RT range-time image.

[0007] The sea surface RT range-time image in the present invention is a range and pulse dimension image obtained after receiving and demodulating the pulse Doppler radar; As an optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation characteristics, in the step 1), the autocorrelation function of the sea surface RT range-time image is: ; Where, M is the number of distance units of the sea surface RT range-time image; N is the number of pulses of the sea surface RT range-time image; represents the echo pixel value of the distance unit m in the nth pulse; p represents the number of pulse intervals substituted into the formula calculation; The constructed autocorrelation matrix is: ; Where K is the number of time intervals in which the autocorrelation function converges to 0.

[0008] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation characteristics, the operation of constructing the autocorrelation matrix from the autocorrelation function is: Due to the signal characteristics of sea clutter, the autocorrelation function converges to 0, and the time interval K for the autocorrelation function to converge to 0 is obtained by the following formula: ; In the formula, Indicates a threshold close to 0, taking 0.01; If the autocorrelation function values ​​of the time intervals greater than K are set to zero, the autocorrelation matrix formed by the autocorrelation functions of the sea clutter data is a Toeplitz matrix, that is, an autocorrelation matrix.

[0009] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation characteristics, in the step 2), the method of constructing the covariance matrix of the speckle component is: 2.1) Use Gaussian distribution as the starting sample to generate sample noise, and then construct a complex Gaussian white noise matrix W; 2.2) The complex Gaussian white noise matrix is ​​transformed into a colored noise matrix with given correlation characteristics, thereby generating a speckle component reflecting the correlation information of sea clutter; 2.3) The autocorrelation matrix of sea clutter is used to represent the covariance matrix of the speckle component.

[0010] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation characteristics, in 2.1), the complex Gaussian white noise matrix W is: ; In the formula, and is an M×N order Gaussian white noise matrix.

[0011] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation characteristics, the colored noise matrix in 2.2) is: ; Where W is the complex Gaussian white noise matrix, and L is the lower triangular matrix obtained by Cholesky decomposition of the autocorrelation matrix C.

[0012] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation characteristics, the covariance matrix of the speckle component in 2.3) is: ; Where R is the covariance matrix of the speckle component, represents the colored noise matrix, C is the autocorrelation matrix of sea clutter, , L is the lower triangular matrix obtained by Cholesky decomposition of the correlation matrix C, L H is the transpose of L, W is the complex Gaussian white noise matrix, W H is the transpose of W.

[0013] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation characteristics, the specific operation of obtaining the eigenvector matrix of the speckle component in step 3) is: The covariance matrix of the speckle component is decomposed by eigenvalue, and the covariance matrix, eigenvalue and eigenvector of the speckle component satisfy the following relationship: ; Where R is the covariance matrix of the speckle component, is a diagonal matrix composed of the eigenvalues ​​of the speckle component covariance matrix; is a matrix composed of the eigenvectors of the speckle component covariance matrix.

[0014] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation features, in step 4), the specific operation of obtaining the subspace of sea clutter is: 4.1) The covariance matrix of the sea surface RT range-time image Unitary transformation is performed to separate the sea clutter and the noise disk, and the separated data covariance matrix is ​​divided into blocks to obtain a covariance matrix with a dimension of (N-1)×(N-1) ; Among them, the separated data covariance matrix is ​​divided into blocks, and its expression is: ; In the formula, is a (N-1)-dimensional column vector; for The conjugate transpose of is also a (N-1)-dimensional row vector; For a single element; 4.2) Obtaining the N-1 dimensional covariance matrix The characteristic matrix , the unitary transformation matrix is ; Where: a is a (N-1)-dimensional column vector whose elements are all 0; b is a (N-1)-dimensional row vector whose elements are all 0; At this time, the covariance matrix after unitary transformation is expressed as ; 4.3) According to the GDE theorem, corresponds to The radius of the Gale circle is the number of feature vectors in the sea surface RT distance-time image. The estimated empirical formula is ; In the formula, Indicates that it corresponds to The radius of the Gale circle, k represents the kth Gale circle, It represents the adjustment factor related to the matrix T, and its purpose is to adjust the mean value of the radius according to the size of the matrix T; 4.4) Obtain the subspace expression of sea clutter: ; in, for Before A matrix of vectors.

[0015] As another optimization scheme of the above-mentioned sea clutter suppression method based on autocorrelation features, in step 5), the specific operation of suppressing sea clutter in the sea surface RT range-time image according to the orthogonal projection operator is: 5.1) According to the projection theorem, the projection operator for orthogonally projecting the sea surface RT range-time image into the subspace of sea clutter is: ; in, represents the subspace of sea clutter; 5.2) Let x be the original sea surface RT distance-time image, is the projected sea surface RT distance-time image, then, the sea surface RT distance-time image after orthogonal projection suppression is expressed as: ; Where: is an orthogonal projection operator, through which the sea clutter in the sea surface RT range-time image is suppressed.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1) The method of the present invention does not need to rely on reference units to construct a covariance matrix, but generates the speckle component of sea clutter through the autocorrelation characteristics in the sea surface RT range-time image, and uses the speckle component to suppress the sea clutter. This method effectively avoids the influence of the texture component of the sea clutter and the target in the sea surface RT range-time image, and provides a new idea for suppressing sea clutter in the sea surface RT range-time image. This innovation not only simplifies the sea clutter suppression process, but also provides new ideas and methods for related fields; 2) The present invention obtains the covariance matrix of the speckle component through the autocorrelation function of the sea surface RT range-time image, determines the number of eigenvectors by using the Gauss circle method, and selects eigenvectors from the covariance matrix of the speckle component to construct the characteristic subspace of the sea clutter. Since the small maneuvering target has little influence on the number of sea clutter eigenvectors, the number of eigenvectors obtained by using the sea surface RT range-time image is close to the true number of pure sea clutter eigenvectors, and since the target has little influence on the sea clutter autocorrelation function, the subspace of the sea clutter obtained by this method will be more accurate. 3) The present invention suppresses sea clutter by orthogonal projection method according to the subspace characteristics of sea clutter, so it can match the Doppler information of sea clutter to effectively suppress sea clutter. When the target moves at the same speed as the sea surface, it will also be suppressed to a certain extent. Therefore, the method proposed by the present invention can also adaptively suppress floating targets in the sea surface RT range-time image, thereby improving the recognition ability of maneuvering targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A graph showing the change in the normalized intensity of Shanghai clutter in the frequency dimension in the sea surface RT range-time image before and after the method of the present invention is used; Figure 2 Schematic diagram of the pixel value amplitude distribution in the sea surface RT range-time image and the pixel value amplitude distribution after being suppressed by the method of the present invention; Figure 3 The unprocessed RT distance-time image of the sea surface and the processed RT distance-time image of the sea surface by the method of the present invention are obtained after a moving target is added to the sea clutter sample; Figure 4 In order to perform incoherent accumulation of 300 pulses, when the target is at 4m / s and SIR is -10, the method of the present invention is used to suppress the image of the distance dimension fluctuation obtained by performing incoherent accumulation of the front and rear sea surface distance-time images in the time dimension; Figure 5In order to perform incoherent accumulation of 300 pulses, when the target is at 4m / s and SIR is -5, the method of the present invention is used to suppress the image of the distance dimension fluctuation obtained by performing incoherent accumulation of the front and rear sea surface distance-time images in the time dimension; Figure 6 In order to perform incoherent accumulation on 300 pulses, when the target is at 4 m / s and SIR is 0, the method of the present invention is used to suppress the image of the distance dimension fluctuation obtained by performing incoherent accumulation of the front and rear sea surface distance-time images in the time dimension; Figure 7 In order to perform incoherent accumulation on 300 pulses, when the target is at 4 m / s and the SIR is 5, the method of the present invention is used to suppress the image of distance dimension fluctuation obtained by performing incoherent accumulation of the front and rear sea surface distance-time images in the time dimension. DETAILED DESCRIPTION

[0018] The method of the present invention is further described in detail below in conjunction with specific embodiments. The parts of the present invention not described in the following embodiments are regarded as prior art known or should be known to those skilled in the art.

[0019] Example 1

[0020] A sea clutter suppression method based on autocorrelation features is proposed. The method is based on a sea surface RT range-time image. First, the autocorrelation function of the sea surface RT range-time image is calculated, and an autocorrelation matrix is ​​constructed based on the function. Then, the covariance matrix of the speckle component is constructed according to the autocorrelation matrix. Subsequently, the covariance matrix of the speckle component is subjected to eigenvalue decomposition to extract its eigenvector matrix. Then, the subspace of the sea clutter is determined according to the number of eigenvectors of the sea surface RT range-time image. Finally, the obtained sea clutter subspace is substituted into an orthogonal projection suppression algorithm to construct an orthogonal projection operator, and the operator is used to effectively suppress the sea clutter in the sea surface RT range-time image. The specific operations are as follows: 1) Obtain the autocorrelation function of the sea surface RT range-time image and construct the autocorrelation matrix; The sea clutter signal z satisfies the wide-sense stationary property. The autocorrelation function is calculated along the time dimension for each distance unit in the sea surface RT range-time image, and the results of all distance units are averaged to obtain the autocorrelation function of the sea surface RT range-time image. The expression of the autocorrelation function is: ; Where: M is the number of distance units of the sea surface RT distance-time image; N is the number of pulses of the sea surface RT distance-time image; represents the echo pixel value of the distance unit m in the nth pulse; p represents the number of pulse intervals substituted into the formula calculation; The autocorrelation function of sea clutter data will slowly converge to 0 as the number of time intervals increases. This is because the signal characteristics of sea clutter usually have strong time variability. The autocorrelation function describes the similarity of signals at different time points. As the time interval increases, the similarity will weaken and eventually approach zero. Therefore, K represents the number of time intervals for the autocorrelation function to converge to 0, which is obtained by the following formula: ; in, Indicates a threshold close to 0, taking 0.01; The autocorrelation function values ​​of the time intervals greater than K are set to zero, and the autocorrelation matrix formed by the autocorrelation function of the sea clutter data is a Toeplitz matrix, that is, an autocorrelation matrix, which can be expressed as: ; Where K represents the number of time intervals that the autocorrelation function converges to 0, and C is an N×N symmetric and positive definite matrix. Since the autocorrelation function of sea clutter data converges to 0, there are many 0 elements in C.

[0021] 2) Construct the covariance matrix of the speckle component based on the autocorrelation matrix; The autocorrelation matrix C can describe the linear relationship between these random variables. Since the Cholesky decomposition can introduce autocorrelation characteristics into independent and identically distributed white noise, the Cholesky decomposition can be used to generate a Gaussian white noise sequence with a specific autocorrelation structure. The Cholesky decomposition can decompose the positive definite matrix C into a lower triangular matrix and its transpose: ; Among them, L is the lower triangular matrix obtained by Cholesky decomposition of C, L H is the transpose of L; According to SIRP theory, the speckle component comes from the echo signals of a large number of basic scatterers, so its characteristics are not affected by the distribution. Since the Gaussian distribution has the excellent property of maintaining the Gaussian characteristics after linear transformation, the Gaussian distribution is selected as the starting sample for generating the correlated sample clutter. Assume and is an M×N order Gaussian white noise matrix, and Construct the complex Gaussian white noise matrix: ; Multiplying the complex white noise matrix with the matrix L obtained from the Cholesky decomposition yields a colored noise matrix with the desired correlation: ; Where W is the complex Gaussian white noise matrix, L is the lower triangular matrix obtained by Cholesky decomposition of the autocorrelation matrix C; Calculate the covariance matrix of the generated speckle components: ; Where R is the covariance matrix of the speckle component, represents the colored noise matrix, C is the autocorrelation matrix of sea clutter, , L is the lower triangular matrix obtained by Cholesky decomposition of the correlation matrix C, L H is the transpose of L, W is the complex Gaussian white noise matrix, W H is the transpose of W; At this point, the covariance matrix of the speckle component can be represented by the autocorrelation matrix of the sea clutter; 3) Obtain the eigenvector matrix of the speckle component; The covariance matrix of the speckle component is decomposed by eigenvalue, and the covariance matrix, eigenvalue and eigenvector of the speckle component satisfy the following relationship: ; Where R is the covariance matrix of the speckle component, is a diagonal matrix composed of the eigenvalues ​​of the speckle component covariance matrix; is a matrix composed of the eigenvectors of the speckle component covariance matrix; 4) Obtain the subspace of sea clutter based on the number of eigenvectors of the sea surface RT range-time image; The sea clutter covariance matrix describes the correlation and variance of sea clutter data in different dimensions, where U can be used as the principal component direction of the sea clutter covariance matrix, which directly reflects the correlation characteristics of the data in these directions, and Indicates the variance in the corresponding direction. The amplitude information of sea clutter is contained in , while U only has the correlation information of sea clutter.

[0022] Since the eigenvectors of the covariance matrices of the speckle component and sea clutter are the same, the eigenvectors corresponding to the large eigenvalues ​​in the covariance matrix of the speckle component can also constitute the subspace of the sea clutter. However, the eigenvalues ​​of the covariance matrices of the speckle component and sea clutter are not consistent, and the number of eigenvectors determined only by the speckle component is inaccurate.

[0023] To solve this problem, the present invention obtains the covariance matrix of sea clutter by means of the sea surface RT distance-time image, and uses the Gerschgorin disk estimation (GDE) method to determine the number of eigenvalue vectors of sea surface echoes. Usually, there is no obvious difference between the sea surface echo Gale circle and the thermal noise Gale circle of the sea clutter covariance matrix. It is necessary to perform a certain transformation on the covariance matrix according to the Gale circle theorem, so that the radius of the sea surface echo Gale circle of the transformed covariance matrix is ​​significantly larger than the radius of the thermal noise Gale circle, and then a reliable estimation of the number of eigenvectors of sea clutter is achieved according to the Gale circle radius.

[0024] The covariance matrix of the sea surface RT distance-time image is Unitary transformation is performed to separate the sea clutter and noise disks. Usually, the disk radius corresponding to sea clutter is larger, while the disk radius corresponding to noise is smaller. The separated data covariance matrix is ​​divided into blocks to obtain a covariance matrix with a dimension of (N-1)×(N-1) , the expression of the block is: ; in, The covariance of the sea surface RT range-time image is obtained, which is the basic knowledge for those skilled in the art; right After block division, The dimension is (N-1)×(N-1); is a (N-1)-dimensional column vector; for The conjugate transpose of is an (N-1)-dimensional row vector; For a single element.

[0025] Get the N-1 dimensional covariance matrix The characteristic matrix , the unitary transformation matrix is: ; Where a is an (N-1)-dimensional column vector whose elements are all 0; b is an (N-1)-dimensional row vector whose elements are all 0.

[0026] Therefore, the covariance matrix after unitary transformation can be expressed as: ; According to the GDE theorem, corresponds to The radius of the Gale circle, according to the distribution characteristics of the Gale circle radius, the number of eigenvectors in the sea surface RT distance-time image The estimated empirical formula is: ; In the formula, Indicates that it corresponds to The radius of the Gale circle, k represents the kth Gale circle, Represents the adjustment factor related to the matrix T. Its main function is to adjust the mean of the radius according to the size of the matrix T.

[0027] Therefore, the subspace of sea clutter is expressed as: ; in, for Before A matrix of vectors 5) Orthogonal projection to suppress sea clutter; The orthogonal projection method can effectively suppress sea clutter. In the orthogonal projection operation, the data will be projected in a specific direction. The present invention obtains the subspace of sea clutter by constructing the covariance matrix of the speckle component, and according to the projection theorem, the projection operator can be obtained as: ; in, represents the subspace of sea clutter; Let x represent the original sea surface RT distance-time image, is the projected sea surface RT distance-time image, then, the sea surface RT distance-time image after orthogonal projection suppression is expressed as: ; Where: is an orthogonal projection operator (i.e., an operator obtained after orthogonal projection operation), by which the sea clutter in the sea surface RT range-time image can be suppressed.

[0028] Example verification: In order to verify the performance of the sea clutter suppression method proposed in the present invention, the data used is from the 2020 No. 1 data file "20210106155330_01_staring" in the "Sea Radar Detection Data Sharing Program (SDRDSP)". This set of data is targetless sea surface observation data with a range resolution of 6 meters and an HH polarization mode. The changes in sea clutter before and after the sea clutter suppression method is used are compared.

[0029] 1. Analysis of sea clutter suppression effect Since the sea clutter suppression algorithm of the present invention obtains an orthogonal projection operator based on the autocorrelation matrix, and there is a Fourier transform correspondence between the autocorrelation function and the power spectrum, the present invention can suppress sea clutter more accurately in the frequency dimension. In order to demonstrate the effect of the present invention on sea clutter suppression, the sea surface observation data in the data file "20210106155330_01_staring" is continued to be used for sea clutter suppression experiments. Figure 1 The change of the normalized intensity of Shanghai clutter in the frequency dimension before and after the application of the method of the present invention is demonstrated.

[0030] Since the sea surface moves with the sea breeze, this causes the echo signal of the sea surface to shift in frequency, so the peak position of the sea clutter power spectrum reflects the frequency shift caused by the movement of the sea surface. Figure 1 It can be seen that the peak value of the power spectrum of sea clutter is near 1, and the peak of the power spectrum of sea clutter after being processed by the present invention is significantly weakened. This is because the autocorrelation function can effectively describe the complex signal form of sea clutter, and the present invention can adaptively match the autocorrelation characteristics of sea clutter, filter a large number of echoes from the moving sea surface, and thus reduce the intensity of sea clutter.

[0031] Figure 2 2 shows the pixel value amplitude distribution of the original sea surface RT distance-time image and the pixel value amplitude distribution after applying the method of the present invention.

[0032] from Figure 2 It can be seen that the tailing feature of the pixel value amplitude distribution image processed by the method of the present invention is reduced, which indicates that the sea peak echo in the moving state is effectively suppressed. Since the sea peak echo has a great influence on the target detection performance, suppressing these echoes can significantly improve the target detection effect.

[0033] 2. Improvement of target detection performance by the suppression method of the present invention In order to ensure the reliability of the experimental results, public sea surface observation data was used for comparison. Specifically, 300 consecutive pulses were randomly selected from the 2000 to 2200 distance units in the file "20210106155330_01_staring" as sea clutter samples. A moving target was added to the sea clutter sample, the signal-to-clutter ratio of the target was set to -5, and it moved towards the observation device at a speed of 4m / s. Figure 3 The unprocessed sea surface RT distance-time image and the sea surface RT distance-time image after the data is processed by the method of the present invention are given in FIG. exist Figure 3The target exists in the 101st distance sample in the image. It can be found through observation that the sea clutter completely covers the target, and it is difficult to identify the target with the naked eye. However, after being processed by the method of the present invention, the amplitude distribution of the sea clutter is significantly improved and becomes more uniform, and the target can also be clearly observed. This is because the present invention effectively filters a large number of echoes from the moving sea surface, thereby improving the recognition of targets with different moving speeds from the sea surface.

[0034] Incoherent accumulation is performed on the 300 pulses to verify the improvement of the target detection performance by the present invention. Figure 4-Figure 7 The images of the signal to sea clutter power ratio (SIR) of the target at 4 m / s are -10, -5, 0, and 5 respectively.

[0035] from Figure 4-Figure 7 It is not difficult to find that after being processed by the method of the present invention, the target can be clearly observed in the distance dimension when SIR=-5, while the pulse accumulation image without clutter suppression can only clearly observe the target when SIR=0. It can be seen that the method of the present invention can effectively suppress sea clutter data.

Claims

1. A sea clutter suppression method based on autocorrelation characteristics, characterized in that: The steps include: 1) Calculate the autocorrelation function of each distance unit in the sea surface RT distance-time image along the time dimension, and average the results of all distance units to obtain the autocorrelation function of the sea surface RT distance-time image, and then construct the autocorrelation matrix based on the autocorrelation function of the sea surface RT distance-time image; 2) Construct the covariance matrix of the speckle component based on the autocorrelation matrix; 3) Perform eigenvalue decomposition on the covariance matrix of the speckle component to obtain the eigenvector matrix of the speckle component; 4) Obtain the subspace of sea clutter based on the number of eigenvectors of the sea surface RT range-time image; 5) Substitute the subspace of sea clutter into the orthogonal projection suppression algorithm to obtain the orthogonal projection operator, and use the orthogonal projection operator to effectively suppress the sea clutter in the sea surface RT range-time image.

2. The sea clutter suppression method based on autocorrelation characteristics according to claim 1, characterized in that: In the step 1), the autocorrelation function of the sea surface RT distance-time image is: ； Where, M is the number of distance units of the sea surface RT range-time image; N is the number of pulses of the sea surface RT range-time image; represents the echo pixel value of the distance unit m in the nth pulse; p represents the number of pulse intervals substituted into the formula calculation; The constructed autocorrelation matrix is: ； Where K is the number of time intervals in which the autocorrelation function converges to 0.

3. The sea clutter suppression method based on autocorrelation characteristics according to claim 2 is characterized in that: The operation of constructing the autocorrelation matrix from the autocorrelation function is: Due to the signal characteristics of sea clutter, the autocorrelation function converges to 0, and the time interval K for the autocorrelation function to converge to 0 is obtained by the following formula: ； In the formula, Indicates a threshold close to 0, taking 0.01; If the autocorrelation function values ​​of the time intervals greater than K are set to zero, the autocorrelation matrix formed by the autocorrelation functions of the sea clutter data is a Toeplitz matrix, that is, an autocorrelation matrix.

4. The sea clutter suppression method based on autocorrelation characteristics according to claim 1 is characterized in that: In step 2), the method for constructing the covariance matrix of the speckle component is: 2.1) Use Gaussian distribution as the starting sample to generate sample noise, and then construct a complex Gaussian white noise matrix W; 2.2) The complex Gaussian white noise matrix is ​​transformed into a colored noise matrix with given correlation characteristics, thereby generating a speckle component reflecting the correlation information of sea clutter; 2.3) The autocorrelation matrix of sea clutter is used to represent the covariance matrix of the speckle component.

5. The sea clutter suppression method based on autocorrelation characteristics according to claim 4 is characterized in that: In 2.1), the complex Gaussian white noise matrix W is: ； In the formula, and is an M×N order Gaussian white noise matrix.

6. The sea clutter suppression method based on autocorrelation characteristics according to claim 4 is characterized in that: The colored noise matrix in 2.2) is: ； Where W is the complex Gaussian white noise matrix, and L is the lower triangular matrix obtained by Cholesky decomposition of the autocorrelation matrix C.

7. The sea clutter suppression method based on autocorrelation characteristics according to claim 4 is characterized in that: The covariance matrix of the speckle component in 2.3) is: ； Where R is the covariance matrix of the speckle component, represents the colored noise matrix, C is the autocorrelation matrix of sea clutter, , L is the lower triangular matrix obtained by Cholesky decomposition of the correlation matrix C, L H is the transpose of L, W is the complex Gaussian white noise matrix, W H is the transpose of W.

8. The sea clutter suppression method based on autocorrelation characteristics according to claim 1, characterized in that: In step 3), the specific operation of obtaining the eigenvector matrix of the speckle component is: The covariance matrix of the speckle component is decomposed by eigenvalue, and the covariance matrix, eigenvalue and eigenvector of the speckle component satisfy the following relationship: ； Where R is the covariance matrix of the speckle component, is a diagonal matrix composed of the eigenvalues ​​of the speckle component covariance matrix; is a matrix composed of the eigenvectors of the speckle component covariance matrix.

9. The sea clutter suppression method based on autocorrelation characteristics according to claim 1, characterized in that: In step 4), the specific operation of obtaining the subspace of sea clutter is: 4.1) The covariance matrix of the sea surface RT range-time image Unitary transformation is performed to separate the sea clutter and the noise disk, and the separated data covariance matrix is ​​divided into blocks to obtain a covariance matrix with a dimension of (N-1)×(N-1) ; Among them, the separated data covariance matrix is ​​divided into blocks, and its expression is: ； In the formula, is a (N-1)-dimensional column vector; for The conjugate transpose of is also a (N-1)-dimensional row vector; For a single element; 4.2) Obtaining the N-1 dimensional covariance matrix The characteristic matrix , the unitary transformation matrix is ； Where: a is a (N-1)-dimensional column vector whose elements are all 0; b is a (N-1)-dimensional row vector whose elements are all 0; At this time, the covariance matrix after unitary transformation is expressed as ； 4.3) According to the GDE theorem, corresponds to The radius of the Gale circle is the number of feature vectors in the sea surface RT distance-time image. The estimated empirical formula is ； In the formula, Indicates that it corresponds to The radius of the Gale circle, k represents the kth Gale circle, It represents the adjustment factor related to the matrix T, and its purpose is to adjust the mean value of the radius according to the size of the matrix T; 4.4) Obtain the subspace expression of sea clutter: ； in, for Before A matrix of vectors.

10. The sea clutter suppression method based on autocorrelation characteristics according to claim 1, characterized in that: In step 5), the specific operation of suppressing sea clutter in the sea surface RT range-time image according to the orthogonal projection operator is: 5.1) According to the projection theorem, the projection operator for orthogonally projecting the sea surface RT range-time image into the subspace of sea clutter is: ； in, represents the subspace of sea clutter; 5.2) Let x be the original sea surface RT distance-time image, is the projected sea surface RT distance-time image, then, the sea surface RT distance-time image after orthogonal projection suppression is expressed as: ； Where: is an orthogonal projection operator, through which the sea clutter in the sea surface RT range-time image is suppressed.

Images