Sea surface weak target detection method and system based on sea clutter suppression, and storage medium

Abstract

The invention discloses a sea surface weak target detection method and system based on sea clutter suppression and a storage medium, relates to the field of sea surface target detection, and aims to solve the problems that an existing sea surface target detection method is difficult to effectively suppress sea clutters and is difficult to recognize sea surface weak targets under complex sea conditions. Comprising the following steps: step 1, carrying out wavelet transform on a target echo signal of a radar, and changing the resolution of time and frequency through scale scaling to obtain time-frequency domain components under different scales; step 2, estimating clutters of a distance unit to be detected by using the clutters of the adjacent distance units according to the correlation between the clutters of the adjacent distance units of the sea surface and between multiple observation values of echoes of the same distance resolution unit; step 3, constructing a Hankel matrix for the echo signal, performing singular value decomposition, and eliminating a large singular value corresponding to the sea clutter according to the characteristic that the sea clutter energy in the same Chirp signal is greater than the signal energy to obtain a new Hankel matrix; and 4, echo signal reconstruction is carried out, and a sea surface target is detected.

Description

Technical Field

[0001] The present invention relates to the technical field of sea surface target detection, and in particular to a sea surface weak target detection method, system and storage medium based on sea clutter suppression. Background Art

[0002] In recent years, my country's coastal defense capabilities have been greatly improved, and great progress has been made in ship-borne and ship-borne radar detection technology. Under the new situation and background, small-scale proximity non-cooperative unmanned boats (less than 2m) and drifting mines threaten the navigation safety of ships and fleets on the sea surface; in addition, small reefs, buoys, driftwood, buoys, etc. on the sea surface also affect the safety of our manned / unmanned operating ships and sea rescue ships. Such targets have the characteristics of small radar scattering cross-section, weak reflection echo, slow movement speed, and easy to be submerged by sea clutter with the ups and downs of the waves. Conventional optical and infrared sensors are difficult to detect them in environments such as waves, rain and fog. These weak targets on the sea surface often put forward higher requirements on the sensitivity and reliability of sensor detection, and also make marine detection technology face greater difficulties.

[0003] Weak target detection on the sea surface based on sea clutter suppression plays a vital role in ocean detection. Due to the complexity and dynamics of the marine environment, the presence of sea clutter greatly affects the detection and identification of sea surface targets. Sea clutter mainly includes echo signals generated by natural factors such as waves, wind, and rain. These signals will interfere with the capture of real targets by radar or other detection equipment, thereby reducing detection accuracy. Under high sea conditions, sea clutter signals are often similar to target signals, and traditional detection methods are difficult to effectively distinguish, resulting in missed or false detection of targets. Therefore, the development of efficient sea surface weak target detection methods based on sea clutter suppression has become a research focus and difficulty in the current field of ocean detection. These methods not only need to accurately extract target signals under strong clutter backgrounds, but also improve real-time processing capabilities while ensuring detection accuracy to cope with complex and changing marine environments. Summary of the invention

[0004] The technical problems to be solved by the present invention are:

[0005] Existing sea surface target detection methods are difficult to effectively suppress sea clutter and difficult to identify weak sea surface targets in complex sea conditions.

[0006] The present invention adopts the following technical solutions to solve the above technical problems:

[0007] The present invention provides a method for detecting weak targets on the sea surface based on sea clutter suppression, comprising the following steps:

[0008] Step 1: Perform wavelet transform on the target echo signal of the radar, change the resolution of time and frequency by scaling, obtain time-frequency domain components at different scales, and filter out the noise in the echo signal;

[0009] Step 2: The echo signals after noise filtering are combined into a signal space, and the clutter of the adjacent range unit and the correlation between multiple observation values ​​of the echo of the same range resolution unit are used to estimate the clutter of the range unit to be detected by using the clutter of the adjacent range unit, so as to preliminarily suppress the sea clutter;

[0010] Step 3: construct a Hankel matrix with the echo signal obtained in step 2, and perform singular value decomposition on the Hankel matrix to obtain a singular value matrix. Then, according to the characteristic that the energy of sea clutter is greater than the signal energy in the same Chirp signal, the large singular values ​​corresponding to the sea clutter are eliminated to obtain a new Hankel matrix.

[0011] Step 4: reconstruct the echo signal of the new Hankel matrix obtained in step 3, and finally detect the sea surface target based on the reconstructed echo signal.

[0012] Furthermore, in step 1, the target echo signal of the radar is subjected to wavelet transformation, specifically:

[0013]

[0014] in, is the wavelet basis function, through the mother function Translation and extension, represents complex conjugate, a and b are scale factor and translation factor respectively, which can be used to adjust the size of time window and frequency window, and a≠0;

[0015] The inverse wavelet transform in step 1 is:

[0016]

[0017] in, ψ(ω) is FFT transform.

[0018] Furthermore, in step one, cubic wavelet decomposition is used.

[0019] Furthermore, the step 2 of estimating the clutter of the to-be-detected distance unit by using the clutter of the adjacent distance unit comprises the following steps:

[0020] Use the clutter of the adjacent range unit to estimate the clutter of the range unit to be detected

[0021]

[0022] and

[0023]

[0024] Among them, the radar echo signal x of the i-th range unit i =[x i (1),x i (2),…,x i (N)] T , N is the number of observations, superscript T is transposition, α i is the weighting factor, 2L is the number of reference distance units, span{x1,x2,…,x 2L} is a subspace of the signal space;

[0025] From the perspective of vector and projection, in order to minimize the clutter estimation error, As a subspace span{x1,x2,…,x 2L}, that is:

[0026]

[0027] Among them, P x is the projection operator, X r For an N×2L matrix:

[0028] X r =[x1 x2 …x 2L ] N×2L (6)

[0029] α=[α1 α2 …α 2L ] T (7)

[0030] According to the projection theorem:

[0031] <x i ,xP x x>=0 (8)

[0032] Among them, the value range of i is i=1,2,…,2L; from equations (5) to (8), we can get:

[0033]

[0034] Wherein, the superscript * represents the conjugate transpose;

[0035] The clutter estimate of the range unit to be detected is obtained as follows:

[0036]

[0037] Subtract the clutter estimate from the signal of the unit to be detected, and we get:

[0038]

[0039] in, is the orthogonal projection operator, and I is the identity matrix.

[0040] Furthermore, step three includes the following steps:

[0041] The echo signal x of the radar's i-th chirp i It is expressed as:

[0042] x i =[x i (1),x i (2),…,x i (N)] T (12)

[0043] Convert it into a Hankel matrix, expressed as:

[0044]

[0045] Where H is a (N-m+1)×m matrix, and m is the number of columns of the Hankel matrix;

[0046] Perform singular value decomposition on the matrix H:

[0047]

[0048] Where U and V are the (N-m+1)×(N-m+1) and m×m unitary matrices of the matrix H, respectively. V T represents the transpose of the matrix V, δ is the singular value matrix arranged in descending order, specifically the singular value set of the clutter subspace and the target subspace, expressed as:

[0049]

[0050] Among them, β1,β2,…,β k are the singular values ​​of the matrix H, β i is the i-th singular value in descending order;

[0051] Since the proportion of sea clutter in the same Chirp signal is much greater than that of the weak target to be detected, the singular value representing the clutter subspace is greater than the singular value representing the target subspace. Therefore, by setting the larger singular value to zero, the sea clutter component in the echo data is eliminated, and the components representing the target and noise are retained; that is:

[0052] Assume that δ′ is a singular value matrix arranged in descending order, and the singular values ​​represented by the clutter subspace are the first q values. Then the singular value set after removing the sea clutter component is expressed as:

[0053]

[0054] Construct the new Hankel matrix H':

[0055]

[0056] Furthermore, in step 3, q∈[0.2k,0.5k] is taken, where k is the number of all singular values ​​in the singular value matrix.

[0057] Furthermore, in step 4, the echo signal is reconstructed using the new Hankel matrix, specifically:

[0058]

[0059] Among them, λ n is the number of elements on the auxiliary diagonal of the nth dimension, the value range of n is n=1,2,…,N, and l+j-1=n.

[0060] Furthermore, in step 4, the CA-CFAR detector is used to detect sea surface targets.

[0061] The present invention provides a sea surface weak target detection system based on sea clutter suppression. The system has a program module corresponding to the steps of the method described in any one of the above technical solutions, and executes the steps in the above sea surface weak target detection method based on sea clutter suppression during operation.

[0062] A computer-readable storage medium stores a computer program, wherein the computer program is configured to implement the steps of the method for detecting weak targets on the sea surface based on sea clutter suppression described in any one of the above technical solutions when called by a processor.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] The invention proposes a method for detecting weak targets on the sea surface based on sea clutter suppression. The method preliminarily suppresses the sea clutter by solving the projection operator, and further eliminates the sea clutter component in the echo data by singular value decomposition, utilizing the characteristic that the energy of the sea clutter in all distance units is much greater than the energy of the target to be detected. The method effectively suppresses the interference of the sea surface clutter on the radar or other detection equipment, thereby greatly improving the detection accuracy and efficiency of weak targets.

[0065] The target detection method proposed in the present invention achieves a higher detection probability with a lower algorithm complexity, and realizes more accurate detection of sea surface targets under conditions of low signal-to-noise ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 is a flow chart of a method for detecting weak targets on the sea surface based on sea clutter suppression in an embodiment of the present invention;

[0067] Figure 2 This is a measured scenario in an embodiment of the present invention;

[0068] Figure 3 is an input / output signal-to-noise ratio sea clutter suppression curve diagram in an embodiment of the present invention;

[0069] Figure 4 is an original echo data diagram in an embodiment of the present invention;

[0070] Figure 5 This is a clutter suppression result diagram using a traditional SVD algorithm in an embodiment of the present invention;

[0071] Figure 6 This is a result diagram of the CEP-SVD algorithm based on wavelet transform in an embodiment of the present invention;

[0072] Figure 7 Result diagram of clutter suppression of the CEP-SVD algorithm based on wavelet transform in an embodiment of the present invention;

[0073] Figure 8 Result diagram of the CEP-SVD-OP algorithm based on wavelet transform in an embodiment of the present invention;

[0074] Fig. 9 3 is a diagram of the detection results of CA-CFAR when the false alarm rate Pfa=0.001 in an embodiment of the present invention. DETAILED DESCRIPTION

[0075] In order to enable those skilled in the art to better understand the scheme of the present invention, exemplary implementations or embodiments of the present invention will be described below in conjunction with the accompanying drawings. Obviously, the described implementations or embodiments are only implementations or embodiments of a part of the present invention, not all of them. Based on the implementations or embodiments of the present invention, all other implementations or embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0076] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0077] The present invention provides a method for detecting weak targets on the sea surface based on sea clutter suppression, comprising the following steps:

[0078] Step 1: Perform wavelet transform on the target echo signal of the radar to filter out random noise in the time-frequency domain. The wavelet changes the resolution of time and frequency by scaling to obtain time-frequency domain components at different scales. Perform inverse wavelet transform to obtain the echo signal after noise filtering.

[0079] Step 2: The echo signals after noise filtering are combined into a signal space, and the correlation between the clutter of the adjacent range units and the multiple observation values ​​of the echo of the same range resolution unit is utilized. The clutter of the range unit to be detected is estimated by using the clutter of the adjacent range unit, and the clutter signal in the target signal to be detected is removed, so as to preliminarily suppress the sea clutter.

[0080] Step 3: construct a Hankel matrix with the echo signal obtained in step 2, and perform singular value decomposition on the Hankel matrix to obtain a singular value matrix. Then, according to the characteristic that the energy of sea clutter is greater than the signal energy in the same Chirp signal, the large singular values ​​corresponding to the sea clutter are eliminated to obtain a new Hankel matrix.

[0081] Step 4: reconstruct the echo signal of the new Hankel matrix obtained in step 3, and finally detect the sea surface target based on the reconstructed echo signal.

[0082] Preferably, in step 1, the target echo signal of the radar is subjected to wavelet transformation, specifically:

[0083]

[0084] in, is the wavelet basis function, through the mother function Translation and extension, represents complex conjugate, a and b are scale factor and translation factor respectively, which can be used to adjust the size of time window and frequency window, so as to realize the control of time and frequency resolution, and a≠0. The time-frequency window structure of wavelet transform can meet the basic clutter removal requirements.

[0085] The inverse wavelet transform in step 1 is:

[0086]

[0087] in, ψ(ω) is FFT transform.

[0088] Wavelet transform is a signal conversion from time domain to frequency domain. Wavelet transform can achieve better time resolution in the high-frequency part of the signal and better frequency resolution in the low-frequency part of the signal, thereby separating effective information from the initial radar signal.

[0089] Preferably, in step one, cubic wavelet decomposition is used.

[0090] Preferably, the step 2 of estimating the clutter of the to-be-detected distance unit by using the clutter of the adjacent distance unit comprises the following steps:

[0091] Use the clutter of the adjacent range unit to estimate the clutter of the range unit to be detected

[0092]

[0093] and

[0094]

[0095] Among them, the radar echo signal x of the i-th range unit i =[x i (1),x i (2),…,x i (N)] T , N is the number of echo signals, superscript T is transposition, α i is the weighting factor, 2L is the number of reference distance units, span{x1,x2,…,x 2L} is a subspace of the signal space;

[0096] From the perspective of vector and projection, in order to minimize the clutter estimation error, As a subspace span{x1,x2,…,x 2L}, that is:

[0097]

[0098] Among them, P x is the projection operator, X r For an N×2L matrix:

[0099] X r =[x1 x2 …x 2L ] N×2L (6)

[0100] α=[α1 α2 …α 2L ] T (7)

[0101] According to the projection theorem:

[0102] <x i ,xP x x>=0 (8)

[0103] Among them, the value range of i is i=1,2,…,2L; from equations (5) to (8), we can get:

[0104]

[0105] Wherein, the superscript * represents the conjugate transpose;

[0106] The clutter estimate of the range unit to be detected is obtained as follows:

[0107]

[0108] Subtract the clutter estimate from the signal of the unit to be detected, and we get:

[0109]

[0110] in, is the orthogonal projection operator, and I is the identity matrix.

[0111] This implementation scheme is simple and easy to implement in engineering and has good detection performance.

[0112] The traditional singular value decomposition algorithm process is:

[0113] The radar echo signal x of the i-th range unit i It can be expressed as:

[0114] x i =[x i (1),x i (2),…,x i (N)] T

[0115] This sequence can be converted into a Hankel matrix, expressed as:

[0116]

[0117] Among them, H is a (N-m+1)×m matrix, m is the number of columns of the Hankel matrix. When the value of m is too small, the suppression effect on the target amplitude is strengthened, resulting in a smaller target amplitude. When the value of m is too large, it will lead to negative effects such as strengthening the suppression effect on the target amplitude, increasing the algorithm complexity, and reducing the algorithm efficiency. The singular value decomposition of the matrix H is:

[0118]

[0119] Where U and V are the (N-m+1)×(N-m+1) and m×m unitary matrices of the matrix H, respectively. V T represents the transpose of the matrix V, δ is the singular value matrix arranged in descending order, specifically the singular value set of the clutter subspace and the target subspace, which can be expressed as:

[0120]

[0121] Among them, β1,β2,…,β k are the singular values ​​of the matrix H, β i is the i-th singular value arranged in descending order. The traditional singular value method believes that the first q large singular values ​​contain the main target components and some clutter components. Therefore, the first q large singular value components are retained and the remaining singular values ​​are set to zero:

[0122]

[0123] Get the new Hankel matrix:

[0124]

[0125] In view of the problem that the traditional SVD algorithm has high complexity and high false alarm rate, step three of this embodiment uses the characteristic that the energy of sea clutter in all distance units is much greater than the energy of the target to be detected, and proposes a sea clutter suppression algorithm based on CEP-SVD, including the following steps:

[0126] The echo signal x of the radar's i-th chirp i It is expressed as:

[0127] x i =[x i (1),x i (2),…,x i (N)] T (12)

[0128] Convert it into a Hankel matrix, expressed as:

[0129]

[0130] Where H is a (N-m+1)×m matrix, and m is the number of columns of the Hankel matrix;

[0131] Perform singular value decomposition on the matrix H:

[0132]

[0133] Where U and V are the (N-m+1)×(N-m+1) and m×m unitary matrices of the matrix H, respectively. V T represents the transpose of the matrix V, δ is the singular value matrix arranged in descending order, specifically the singular value set of the clutter subspace and the target subspace, expressed as:

[0134]

[0135] Among them, β1,β2,…,β k are the singular values ​​of the matrix H, βi are the i-th singular values ​​in descending order;

[0136] In order to confirm the singular values ​​representing the target subspace and the clutter subspace, the present invention considers that there are only a limited number of targets in all the distance cells on a Chirp. Since the proportion of sea clutter in the same Chirp signal is much greater than that of the weak target to be detected, the singular value representing the clutter subspace is greater than the singular value representing the target subspace. Therefore, by setting the larger singular value to zero, the sea clutter component in the echo data is eliminated, and the components representing the target and noise are retained; that is:

[0137] Assume that δ′ is a singular value matrix arranged in descending order, and the singular values ​​represented by the clutter subspace are the first q values. Then the singular value set after removing the sea clutter component is expressed as:

[0138]

[0139] Construct the new Hankel matrix H':

[0140]

[0141] Preferably, in step 3, q∈[0.2k,0.5k] is taken, where k is the number of all singular values ​​in the singular value matrix. In this embodiment, selecting a suitable demarcation point q value can better distinguish the target from the sea clutter, more effectively suppress the sea clutter, and then detect the target from the sea clutter.

[0142] Preferably, in step 4, the echo signal is reconstructed using the new Hankel matrix, specifically:

[0143]

[0144] Among them, λ n is the number of elements on the auxiliary diagonal of the nth dimension, the value range of n is n=1,2,…,N, and l+j-1=n.

[0145] Preferably, in step 4, a CA-CFAR detector is used to detect sea surface targets.

[0146] The method (algorithm) for detecting weak targets on the sea surface based on sea clutter suppression proposed in the present invention is the underlying technical core of the present invention, and various products can be derived based on the algorithm.

[0147] Based on the method proposed in the present invention, a sea surface weak target detection system based on sea clutter suppression is developed using a programming language. The system has a program module corresponding to the steps of the above technical solution, and executes the steps in the above sea surface weak target detection method based on sea clutter suppression during operation.

[0148] The computer program of the developed system (software) is stored on a computer-readable storage medium, and the computer program is configured to implement the steps of the above-mentioned sea surface weak target detection method based on sea clutter suppression when called by a processor, that is, the present invention is materialized on a carrier to become a computer program product.

[0149] Various implementations of the systems and techniques described herein can be realized in digital electronic circuit systems, integrated circuit systems, dedicated ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0150] The computer programs (also referred to as programs, software, software applications, or codes) of the present invention include machine instructions for programmable processors, and these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or device (e.g., disk, optical disk, memory, programmable logic device PLD) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0151] Example 1

[0152] A self-tested sea surface target data set is used, which is a stationary sea surface target, to analyze the detection performance differences among the traditional singular value decomposition algorithm (SVD) algorithm based on wavelet transform, the energy priority-singular value decomposition (CEP-SVD) algorithm, the orthogonal projection (OP) algorithm and the CEP-SVD-OP algorithm of the present invention. Figure 2 As shown in the figure, the test site is located inside the Yuanyao Fishing Port in Weihai, Shandong. The weak target to be tested on the sea surface is a steel oil drum with a height of 48.5cm and a diameter of 38.8cm. It is about 50m away from the radar. The radar frequency band is 77GHz, the range is 128m, the distance resolution is 0.25m, the sampling frequency is 30MHz, the number of sampling points is 1024, the sweep rise time is 35 microseconds, and the sweep bandwidth is 615MHz. Figure 3It can be seen that the CEP-SVD algorithm is always effective in the area of ​​input signal-to-noise ratio of -30dB to 10dB, and the signal-to-noise ratio gain is approximately 37dB. The clutter suppression capability of the CEP-SVD-OP algorithm is further improved compared with the CEP-SVD algorithm. The suppression performance of the SVD algorithm and the OP algorithm is poor and cannot adapt to the scene with too strong clutter noise. Collect radar echo signal data ( Figure 4 ), the original signal before processing has strong clutter intensity, small signal-to-noise ratio, and the calculated signal-to-noise ratio is 5.76dB; for the traditional SVD algorithm, the number of columns of the Hankel matrix is ​​m = 50, the order of the singular value is q = 1, and the clutter signal intensity after the traditional SVD algorithm is weakened compared with that before processing, but the target signal intensity is also weakened ( Figure 5 ), the signal-to-noise ratio is calculated to be 7.45dB; for the CEP-SVD algorithm, the number of columns of the Hankel matrix is ​​m = 50, the order of the singular value is q = 4, the signal clutter intensity after the CEP-SVD algorithm is processed is significantly weakened, and the signal-to-noise ratio of the target signal distance unit is improved ( Figure 6 , Figure 7 ), the signal-to-noise ratio is calculated to be 16.64dB; Figure 8 As shown in Figure 2, the signal-to-noise ratio is significantly improved after the CEP-SVD-OP algorithm is processed, and the calculated signal-to-noise ratio is 21.69 dB; Fig. 9 As shown, a CA-CFAR detector (false alarm rate setting pfa=0.001) is used to perform target detection on the data processed by the CEP-SVD-OP algorithm, and the target detection situation is observed and recorded. The green line represents the CA-CFAR (unit average constant false alarm) detection threshold value, and the red line represents the detection target exceeding the threshold. The clutter suppression and signal-to-noise ratio improvement of each algorithm in this embodiment show that the CEP-SVD-OP algorithm has significant advantages in suppressing sea clutter and detecting weak targets.

[0153] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A method for detecting weak targets on the sea surface based on sea clutter suppression, characterized in that: The following steps are involved: Step 1: Perform wavelet transform on the target echo signal of the radar, change the resolution of time and frequency by scaling, obtain time-frequency domain components at different scales, and filter out the noise in the echo signal; Step 2: The echo signals after noise filtering are combined into a signal space, and the clutter of the adjacent range unit and the correlation between multiple observation values ​​of the echo of the same range resolution unit are used to estimate the clutter of the range unit to be detected by using the clutter of the adjacent range unit, so as to preliminarily suppress the sea clutter; Step 3: construct a Hankel matrix with the echo signal obtained in step 2, and perform singular value decomposition on the Hankel matrix to obtain a singular value matrix. Then, according to the characteristic that the energy of sea clutter is greater than the signal energy in the same Chirp signal, the large singular values ​​corresponding to the sea clutter are eliminated to obtain a new Hankel matrix. Step 4: reconstruct the echo signal of the new Hankel matrix obtained in step 3, and finally detect the sea surface target based on the reconstructed echo signal.

2. The method for detecting weak targets on the sea surface based on sea clutter suppression according to claim 1, characterized in that: In step 1, the target echo signal of the radar is transformed by wavelet transform, specifically: in, is the wavelet basis function, through the mother function Translation and extension, represents complex conjugate, a and b are scale factor and translation factor respectively, which can be used to adjust the size of time window and frequency window, and a≠0; The inverse wavelet transform in step 1 is: in, ψ(ω) is FFT transform.

3. The method for detecting weak targets on the sea surface based on sea clutter suppression according to claim 2 is characterized in that: In step 1, cubic wavelet decomposition is used.

4. The method for detecting weak targets on the sea surface based on sea clutter suppression according to claim 1, characterized in that: The step 2 of estimating the clutter of the to-be-detected distance unit by using the clutter of the adjacent distance unit comprises the following steps: Use the clutter of the adjacent range unit to estimate the clutter of the range unit to be detected and Among them, the radar echo signal x of the i-th range unit i =[x i (1),x i (2),…,x i (N)] T , N is the number of observations, superscript T is transposition, α i is the weighting factor, 2L is the number of reference distance units, span{x1,x2,…,x 2L } is a subspace of the signal space; From the perspective of vector and projection, in order to minimize the clutter estimation error, As a subspace span{x1,x2,…,x 2L }, that is: Among them, P x is the projection operator, X r For an N×2L matrix: X r =[x1 x2 … x 2L ] N×2L (6) α＝[α1 α2 …α 2L ] T (7) According to the projection theorem: <x i ,x-P x x>=0 (8) Among them, the value range of i is i=1,2,…,2L; from equations (5) to (8), we can get: Wherein, the superscript * represents the conjugate transpose; The clutter estimate of the range unit to be detected is obtained as follows: Subtract the clutter estimate from the signal of the unit to be detected, and we get: in, is the orthogonal projection operator, and I is the identity matrix.

5. The method for detecting weak targets on the sea surface based on sea clutter suppression according to claim 4 is characterized in that: Step three includes the following steps: The echo signal x of the radar's i-th chirp i It is expressed as: x i =[x i (1),x i (2),…,x i (N)] T (12) Convert it to Hankel matrix, expressed as: Where H is a (N-m+1)×m matrix, and m is the number of columns of the Hankel matrix; Perform singular value decomposition on the matrix H: Where U and V are the (N-m+1)×(N-m+1) and m×m unitary matrices of the matrix H, respectively. V T represents the transpose of the matrix V, δ is the singular value matrix arranged in descending order, specifically the singular value set of the clutter subspace and the target subspace, expressed as: Among them, β1,β2,…,β k are the singular values ​​of the matrix H, β i is the i-th singular value in descending order; Since the proportion of sea clutter in the same Chirp signal is much greater than that of the weak target to be detected, the singular value representing the clutter subspace is greater than the singular value representing the target subspace. Therefore, by setting the larger singular value to zero, the sea clutter component in the echo data is eliminated, and the components representing the target and noise are retained; that is: Assume that δ′ is a singular value matrix arranged in descending order, and the singular values ​​represented by the clutter subspace are the first q values. Then the singular value set after removing the sea clutter component is expressed as: Construct the new Hankel matrix H':

6. The method for detecting weak targets on the sea surface based on sea clutter suppression according to claim 5, characterized in that: In step 3, take q∈[0.2k,0.5k], where k is the number of all singular values ​​in the singular value matrix.

7. The method for detecting weak targets on the sea surface based on sea clutter suppression according to claim 6, characterized in that: In step 4, the echo signal is reconstructed using the new Hankel matrix, specifically: Among them, λ n is the number of elements on the auxiliary diagonal of the nth dimension, the value range of n is n=1,2,…,N, and l+j-1=n.

8. The method for detecting weak targets on the sea surface based on sea clutter suppression according to claim 7, characterized in that: In step 4, the CA-CFAR detector is used to detect sea surface targets.

9. A sea surface weak target detection system based on sea clutter suppression, characterized in that: The system has a program module corresponding to the steps of the method described in any one of claims 1 to 8, and executes the steps of the above-mentioned sea surface weak target detection method based on sea clutter suppression when running.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps in the method for detecting weak targets on the sea surface based on sea clutter suppression according to any one of claims 1 to 8 when called by a processor.

Images