Ship target discrimination method and device based on frequency agile radar

Abstract

The application relates to a ship target identification method and device based on a frequency agile radar, wherein the method comprises the following steps: transmitting a frequency agile waveform by using a radar, obtaining a range-velocity distribution map of a target by using a high-resolution range-velocity processing method, extracting features of the measured range-velocity distribution map and inputting the features into a classification network, so as to effectively classify ship targets and passive jamming false targets, and improve the classification performance of the ship targets and the passive jamming false targets. Therefore, the problems in the existing ship target identification technology that the high-resolution range image features are easily disturbed by a viewing angle, the micro-motion features are too dependent on long-time observation, and the classification performance of the ship targets and the passive jamming false targets is greatly affected are solved.

Description

Technical Field

[0001] This application relates to the field of radar target recognition technology, and in particular to a method and device for ship target identification based on frequency-agile radar. Background Technology

[0002] In the field of ship target identification, high-resolution range profiles (HRRP) and micro-motion features are two key technical means, but each faces specific challenges.

[0003] First, HRRP is a one-dimensional vector formed by superimposing the projections of target scattering points along the radar line of sight. While it can reflect the distribution of target scattering centers, its low feature dimension means that projections of different target classes in a certain direction may have high similarity. This reduces the inter-class distance and increases the risk of misclassification. Furthermore, HRRP is highly sensitive to the observation viewpoint. The HRRP features of the same target under different viewpoints exhibit drastically different scattering center distributions, resulting in large intra-class distances and further increasing the difficulty of classification. Traditional filtering-based observation viewpoint estimation methods have low success rates in complex scenes, such as those with interference, which limits the application scope and accuracy of HRRP-based classification methods.

[0004] Secondly, since each target possesses unique structural dynamics, target identification utilizing the physical uniqueness of its micro-motion characteristics has high accuracy. Here, micro-motion refers to the minute movements of the target and its components, such as vibration and rotation, excluding translational motion. However, micro-motion feature extraction requires long-term observation to accumulate the target's micro-motion periodic information, which is a significant disadvantage in application scenarios with limited radar illumination resources. Furthermore, commonly used time-frequency analysis methods have high computational complexity and demanding hardware resources, limiting their widespread application in practical systems.

[0005] In summary, in existing ship target recognition technologies, high-resolution range image features are easily affected by viewing angle interference, and micro-motion features rely too much on long-term observation, which greatly affects the classification performance of ship targets and passive interference false targets, and urgently needs to be solved. Summary of the Invention

[0006] This application provides a method and apparatus for ship target identification based on frequency-agile radar, in order to solve the problems in existing ship target identification technologies, such as high-resolution range image features being easily affected by viewing angle interference, and micro-motion features relying too much on long-term observation, which greatly affects the classification performance of ship targets and passive interference false targets.

[0007] The first aspect of this application provides a method for ship target identification based on frequency-agile radar, comprising the following steps: transmitting a pulse signal to a target under test through a preset frequency-agile radar to obtain an echo signal corresponding to the target under test; performing pulse compression, Doppler compensation, and range migration compensation on the echo signal; performing inverse Fourier transform on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain a high-resolution range image of the target under test; setting different velocity grid points and repeating the signal processing operation to obtain a series of corresponding high-resolution range images; stitching the series of corresponding high-resolution range images according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target under test; and extracting multiple values ​​from each range-velocity distribution map. The scattering point coordinates of the target scattering point are calculated, and the scattering point intensity and intensity weight corresponding to the scattering point coordinates are calculated. Based on the scattering point intensity and intensity weight corresponding to each target scattering point coordinate, the weighted mean square residual, velocity weighted standard deviation, and distance weighted standard deviation corresponding to each distance-velocity distribution map are calculated. The corresponding complementary contrast factor is calculated through the velocity weighted standard deviation and the distance weighted standard deviation. Based on the weighted mean square residual, the complementary contrast factor, the velocity weighted standard deviation, and the distance weighted standard deviation, feature vectors of all distance-velocity distribution maps are constructed, and the feature vectors are input into a support vector machine to obtain the corresponding rigid body discrimination model. The rigid body discrimination model is used to identify the target to be tested, so as to obtain the identification result corresponding to the target to be tested.

[0008] Optionally, in one embodiment of this application, the step of transmitting a pulse signal to the target under test via a preset frequency-agile radar to obtain the echo signal corresponding to the target under test, performing pulse compression, Doppler compensation, and range migration compensation on the echo signal, and performing inverse Fourier transform on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain a high-resolution range image of the target under test, and setting different velocity grid points, repeating the signal processing operation to obtain a series of corresponding high-resolution range images, and stitching the series of corresponding high-resolution range images according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target under test, includes: transmitting the pulse signal to the target under test via the frequency-agile radar, and equating the target under test to multiple corresponding target scattering points, obtaining the echo signal corresponding to each target scattering point based on a preset "stop-jump model"; and performing pulse compression, Doppler compensation, and range migration compensation on the echo signal, and performing inverse Fourier transform on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain multiple range-velocity distribution maps of the target under test, including: transmitting the pulse signal to the target under test via the frequency-agile radar, and equating the target under test to multiple corresponding target scattering points, and obtaining the echo signal corresponding to each target scattering point based on a preset "stop-jump model"; and performing inverse Fourier transform on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain multiple range-velocity distribution maps of the target under test, including: transmitting the pulse signal to the target under test via the frequency-agile radar, and equating the target under test to multiple corresponding target scattering points; obtaining the echo signal corresponding to each target scattering point; and performing inverse Fourier transform on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain multiple range-velocity distribution maps of the target under test, including: transmitting the pulse signal to the target under test via the frequency-agile radar, and equating the target under test to multiple target scattering points based on a preset "stop-jump model" to obtain the The echo signals are linearly superimposed and demodulated to baseband to obtain the echo signal spectrum corresponding to the target under test. Frequency domain matching processing is performed on the echo signal spectrum, followed by velocity Doppler compensation and range migration compensation. Coherent accumulation is performed on the range migration compensated spectrum, and frequency domain interpolation is used to stitch the coherently accumulated spectrum to unify the spectrum to the same frequency axis. An inverse Fourier transform is then performed on the spectrum unified to the same frequency axis to generate a high-resolution range profile corresponding to the target under test. Based on preset different velocity values, signal processing operations are performed on the echo signals to generate multiple corresponding high-resolution range profiles. These multiple high-resolution range profiles are then stitched together according to the set velocity grid dimension order to obtain the range-velocity distribution map of the target under test within each coherent processing interval of the frequency-agile radar.

[0009] Optionally, in one embodiment of this application, the step of calculating the weighted mean square residual, velocity-weighted standard deviation, and distance-weighted standard deviation corresponding to each distance-velocity distribution map based on the scattering point intensity and intensity weight corresponding to each target scattering point coordinate, and calculating the corresponding complementary contrast factor through the velocity-weighted standard deviation and the distance-weighted standard deviation, includes: performing a linear regression fitting operation on the scattering point coordinates on each distance-velocity distribution map to obtain a corresponding linear regression model; determining the intensity weight corresponding to the scattering point intensity of each target scattering point coordinate, and calculating the weighted mean square residual corresponding to the scattering point coordinate based on the intensity weight and the linear regression model.

[0010] Optionally, in one embodiment of this application, the step of using the rigid body identification model to identify the target to be tested in order to obtain the identification result corresponding to the target to be tested includes: determining whether the target to be tested is a rigid body based on the identification result; when the target type is a rigid body, determining that the target to be tested is a ship target; when the target type is a non-rigid body, determining that the target to be tested is a non-ship false target.

[0011] Optionally, in one embodiment of this application, the mathematical expression for the complementarity contrast factor is:

[0012] in, This represents the complementary contrast factor; This represents the distance-weighted standard deviation; This represents the speed-weighted standard deviation; This represents the numerical stability constant.

[0013] A second aspect of this application provides a ship target identification device based on frequency-agile radar, comprising: a spectrum processing module, used to transmit pulse signals to a target under test through a preset frequency-agile radar to obtain echo signals corresponding to the target under test; perform pulse compression, Doppler compensation, and range migration compensation on the echo signals; and perform inverse Fourier transform on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain a high-resolution range image of the target under test; and set different velocity grid points, repeat the signal processing operation to obtain a series of corresponding high-resolution range images; and stitch the series of corresponding high-resolution range images according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target under test; and a feature extraction module, used to extract multiple features from each range-velocity distribution map of the multiple range-velocity distribution maps. The system calculates the scattering point coordinates of the target scattering point, and the scattering point intensity and intensity weight corresponding to the scattering point coordinates. Based on the scattering point intensity and intensity weight corresponding to each target scattering point coordinate, it calculates the weighted mean square residual, velocity-weighted standard deviation, and distance-weighted standard deviation corresponding to each distance-velocity distribution map, and calculates the corresponding complementary contrast factor through the velocity-weighted standard deviation and the distance-weighted standard deviation. The target recognition module is used to construct feature vectors for all distance-velocity distribution maps based on the weighted mean square residuals, the complementary contrast factor, the velocity-weighted standard deviation, and the distance-weighted standard deviation, and inputs the feature vectors into a support vector machine to obtain the corresponding rigid body discrimination model. The rigid body discrimination model is then used to identify the target to be tested, so as to obtain the identification result corresponding to the target to be tested.

[0014] Optionally, in one embodiment of this application, the spectrum processing module includes: an equivalent unit, used to transmit the pulse signal to the target under test through the frequency-agile radar, and to convert the target under test into multiple corresponding target scattering points, so as to obtain the echo signal corresponding to each target scattering point based on a preset "stop-hop model"; a demodulation unit, used to perform a linear superposition operation on the echo signals corresponding to each target scattering point, and to demodulate the linearly superimposed echo signals to baseband, so as to obtain the echo signal spectrum corresponding to the target under test; and a processing unit, used to perform frequency domain matching processing on the echo signal spectrum, and to perform high-speed processing on the frequency domain matching processed echo signal spectrum. The system performs Doppler compensation and range migration compensation operations, coherently accumulates the spectrum after range migration compensation, and then performs frequency domain interpolation and stitching on the coherently accumulated spectrum to unify the spectrum to the same frequency axis. An inverse Fourier transform is then performed on the spectrum unified to the same frequency axis to generate a high-resolution range image corresponding to the target. A stitching unit performs signal processing operations on the echo signal based on preset different velocity values ​​to generate multiple corresponding high-resolution range images. These multiple high-resolution range images are then stitched together according to the set velocity grid dimension order to obtain the range-velocity distribution map of the target within each coherent processing interval of the frequency-agile radar.

[0015] Optionally, in one embodiment of this application, the feature extraction module includes: a fitting unit, used to perform a linear regression fitting operation on the scattering point coordinates on each distance-velocity distribution map to obtain a corresponding linear regression model; and a calculation unit, used to determine the intensity weight corresponding to the scattering point intensity of each target scattering point coordinate, and calculate the weighted mean square residual corresponding to the scattering point coordinate based on the intensity weight and the linear regression model.

[0016] Optionally, in one embodiment of this application, the target recognition module includes: a judgment unit, configured to determine whether the target to be tested is a rigid body based on the identification result; a first judgment unit, configured to determine that the target to be tested is a ship target when the target type is a rigid body; and a second judgment unit, configured to determine that the target to be tested is a non-ship false target when the target type is a non-rigid body.

[0017] Optionally, in one embodiment of this application, the mathematical expression for the complementarity contrast factor is:

[0018] in, This represents the complementary contrast factor; This represents the distance-weighted standard deviation; This represents the speed-weighted standard deviation; This represents the numerical stability constant.

[0019] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the ship target identification method based on frequency-agile radar as described in the above embodiments.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described ship target identification method based on frequency-agile radar.

[0021] A fifth aspect of this application provides a computer program product, including a computer program that is executed to implement the above-described method for ship target identification based on frequency-agile radar.

[0022] Therefore, the embodiments of this application have the following beneficial effects: In this embodiment, a pulse signal is transmitted to the target using a preset frequency-agile radar to obtain the corresponding echo signal. The echo signal is then subjected to pulse compression, Doppler compensation, and range migration compensation. An inverse Fourier transform is performed on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain a high-resolution range image of the target. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. These high-resolution range images are stitched together according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target. The scattering point coordinates of multiple target scattering points in each of the multiple range-velocity distribution maps are extracted, and the scattering point intensity corresponding to the scattering point coordinates is calculated. Based on the intensity of the scattering point corresponding to the coordinates of each target scattering point and its intensity weight, the weighted mean square residual, velocity-weighted standard deviation, and distance-weighted standard deviation are calculated for each range-velocity distribution map. The corresponding complementary contrast factor is then calculated using the velocity-weighted standard deviation and the distance-weighted standard deviation. Based on these weighted mean square residuals, complementary contrast factors, velocity-weighted standard deviations, and distance-weighted standard deviations, feature vectors for all range-velocity distribution maps are constructed. These feature vectors are then input into a support vector machine to obtain the corresponding rigid body discrimination model. This rigid body discrimination model is used to identify the target to obtain the identification result, thus achieving efficient identification of ship targets and passive interference targets, improving the accuracy and generalization performance. This solves the problems in existing ship target identification technologies, such as the susceptibility of high-resolution range image features to viewpoint interference and the over-reliance on long-term observation for micro-motion features, which significantly affect the classification performance of ship targets and passive interference false targets.

[0023] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a ship target identification method based on frequency-agile radar according to an embodiment of this application; Figure 2 A schematic diagram of the execution logic of a ship target identification method based on frequency-agile radar provided for one embodiment of this application; Figure 3 A schematic diagram of the execution logic of a ship target identification method based on frequency-agile radar provided for one embodiment of this application; Figure 4 A schematic diagram of the execution logic of a ship target identification method based on frequency-agile radar provided for one embodiment of this application; Figure 5 This is an example diagram of a ship target identification device based on frequency-agile radar according to an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0025] Among them, 10-ship target identification device based on frequency agile radar; 100-spectrum processing module, 200-feature extraction module, 300-target recognition module; 601-memory, 602-processor, 603-communication interface. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The following describes a method and apparatus for ship target identification based on frequency-agile radar according to embodiments of this application, with reference to the accompanying drawings. Addressing the problems mentioned in the background art, this application provides a method for ship target identification based on frequency-agile radar. In this method, a pulse signal is transmitted to the target under test using a preset frequency-agile radar to obtain the echo signal corresponding to the target. The echo signal is subjected to pulse compression, Doppler compensation, and range migration compensation. The signal spectrum obtained after coherent accumulation and interpolation stitching is subjected to inverse Fourier transform to obtain a high-resolution range image of the target. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. The series of corresponding high-resolution range images are stitched together according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target. The scattering of multiple target scattering points in each range-velocity distribution map is extracted. The system calculates the coordinates of scattering points and the intensity and weight of the scattering points corresponding to these coordinates. Based on the intensity and weight of the scattering points corresponding to each target's scattering point coordinates, it calculates the weighted mean square residual, velocity-weighted standard deviation, and distance-weighted standard deviation for each range-velocity distribution map. It then calculates the corresponding complementary contrast factor using the velocity-weighted standard deviation and the distance-weighted standard deviation. Based on these weighted mean square residuals, complementary contrast factors, velocity-weighted standard deviations, and distance-weighted standard deviations, it constructs feature vectors for all range-velocity distribution maps. These feature vectors are then input into a support vector machine to obtain the corresponding rigid body discrimination model. This rigid body discrimination model is used to identify the target under test, thus obtaining the identification result for the target. This achieves efficient identification of ship targets and passive interference targets, improving the accuracy and generalization performance of the identification. Therefore, it solves the problems in existing ship target identification technologies, such as the susceptibility of high-resolution range image features to viewpoint interference and the over-reliance on long-term observation for micro-motion features, which greatly affects the classification performance of ship targets and passive interference false targets.

[0028] Specifically, Figure 1 This is a flowchart illustrating a ship target identification method based on frequency-agile radar, provided as an embodiment of this application.

[0029] like Figure 1 As shown, the ship target identification method based on frequency-agile radar includes the following steps: In step S101, a pulse signal is transmitted to the target under test through a preset frequency-agile radar to obtain the corresponding echo signal of the target under test. The echo signal is subjected to pulse compression, Doppler compensation and range migration compensation. The signal spectrum obtained after coherent accumulation and interpolation stitching is subjected to inverse Fourier transform to obtain a high-resolution range image of the target under test. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. The series of corresponding high-resolution range images are stitched together according to the preset velocity dimension order to obtain multiple range and velocity distribution maps of the target under test.

[0030] This application embodiment first transmits agile waveform signals (i.e., pulse signals) through a radar system, and receives the echoes (i.e., echo signals) after being scattered by a sea surface target (i.e., the target under test). Through multi-level signal processing such as echo demodulation, frequency domain matching, Doppler compensation, range migration compensation, and coherent accumulation, a high signal-to-noise ratio one-dimensional range profile of the target is generated. Secondly, this application embodiment can generate a series of HRRPs by setting different velocity values ​​and performing Doppler compensation and other processing operations to address the unique micro-motion characteristics of ship targets (such as ship sway). Finally, the range profiles corresponding to different velocity grid points are fused in a two-dimensional plane to construct a range-velocity distribution map of each scattering point on the target under test.

[0031] It should be noted that this distance-velocity distribution map uses distance as the horizontal axis and velocity as the vertical axis. The intensity of each pixel represents the scattering energy intensity of the corresponding distance-velocity unit, thus intuitively reflecting the motion state of each scattering point on the target.

[0032] Therefore, this embodiment of the application transmits agile waveform signals, receives the corresponding radar echoes after being scattered by the target under test, and combines the target's relative position to the radar to measure its azimuth and radial velocity, thereby synthesizing HRRP and the range-velocity distribution map of the target under test. This enables precise capture of the motion state of the ship's scattering points, providing clear and intuitive data support for sea surface target identification and monitoring, and enhancing the reliability of target detection. Optionally, in one embodiment of this application, a pulse signal is transmitted to the target under test via a preset frequency-agile radar to obtain the echo signal corresponding to the target under test. The echo signal is then subjected to pulse compression, Doppler compensation, and range migration compensation. An inverse Fourier transform is performed on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain a high-resolution range image of the target under test. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. These high-resolution range images are then stitched together according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target under test. This includes: transmitting a pulse signal to the target under test via a frequency-agile radar, and equating the target under test to multiple corresponding target scattering points, obtaining the echo signal corresponding to each target scattering point based on a preset "stop-and-go model"; processing the echo signal corresponding to each target scattering point... The signal is linearly superimposed, and the superimposed echo signal is demodulated to baseband to obtain the echo signal spectrum corresponding to the target under test. Frequency domain matching processing is performed on the echo signal spectrum, followed by velocity Doppler compensation and range migration compensation. Coherent accumulation is performed on the range migration compensated spectrum, and frequency domain interpolation is performed on the coherently accumulated spectrum to unify the spectrum to the same frequency axis. An inverse Fourier transform is then performed on the spectrum unified to the same frequency axis to generate a high-resolution range profile corresponding to the target under test. Based on different preset velocity values, signal processing operations are performed on the echo signal to generate multiple corresponding high-resolution range profiles. These multiple high-resolution range profiles are then stitched together according to the set velocity grid dimension order to obtain the range-velocity distribution map of the target under test within each coherent processing interval of the frequency-agile radar.

[0033] Specifically, such as Figure 2 As shown, embodiments of this application can use a frequency-agile radar to transmit pulse signals. It is assumed that within a coherent processing interval, the frequency-agile radar transmits... The pulse, the first The carrier frequency of each pulse is , No. The transmitted waveform of each pulse can be represented as:

[0034] in, , and These represent the pulse repetition interval and pulse width, respectively; rect(·) represents the square wave function, which is defined as:

[0035] For baseband waveforms, taking the LFM waveform as an example, its specific expression is:

[0036] in, The bandwidth of the LFM waveform; The waveform width.

[0037] When the radar illuminates the ideal point target p, according to the "stop-bounce" model assumption, the first... n The received echo of a pulse can be represented as:

[0038] in, and The complex scattering coefficients of the ideal scattering point p and the complex scattering coefficients for the first scattering point p are respectively. The echo delay of each pulse, and It can be approximated as:

[0039] in, This represents the initial radial distance between the scattering point and the radar. The initial radial velocity of the target scattering point; It is the speed of light.

[0040] In this embodiment, the target to be tested can be equivalent to multiple target scattering points. Therefore, the echo signal obtained by target detection is a linear superposition of the echoes from multiple target scattering points. In this embodiment, the target echo signal can be demodulated to baseband and then subjected to frequency domain matched filtering to obtain the first... The spectrum of the target echo signal for each pulse is shown in the following formula:

[0041] Secondly, embodiments of this application can obtain the following by performing frequency domain matching processing on the echo signal spectrum:

[0042] Furthermore, embodiments of this application can obtain the following after performing Doppler compensation and range migration correction on all pulse echo spectra:

[0043] in, It is the compensation speed value for processing.

[0044] Furthermore, in embodiments of this application, M can be defined as the number of frequency points transmitted by the frequency-agile radar within a coherent processing interval. The spectrum at each frequency point can be obtained through coherent accumulation processing:

[0045] in, Indicates the frequency point used The number of pulses, Indicates the first Frequency points used The signal spectrum of the target echo obtained from the pulse is obtained after frequency domain matching, Doppler compensation and range migration compensation.

[0046] Subsequently, in this embodiment of the application, frequency domain interpolation is performed on the spectrum at each frequency point to unify it to the same frequency axis:

[0047] in, The term represents a uniform frequency axis, and Interp represents linear interpolation.

[0048] Subsequently, in this embodiment of the application, an inverse Fourier transform can be performed on the above formula to obtain the high-resolution range image (HRRP) corresponding to the target under test.

[0049] In summary, for sea surface targets, due to the target's own micro-motion characteristics, the radial velocities of different scattering points on the target will differ. Therefore, the embodiments of this application can set a series of different velocity values. Signal processing operations such as Doppler compensation are performed to generate multiple high-resolution range images. These high-resolution range images are then stitched together to obtain the RV Map of the detected target within a coherent processing interval, as shown in the following formula:

[0050] Joint indicates splicing according to the set speed dimension order.

[0051] Therefore, the embodiments of this application improve the ability to capture target details by generating high-quality range-velocity distribution maps, provide accurate data support for target identification and analysis, and enhance the effectiveness of radar detection.

[0052] In step S102, the coordinates of multiple target scattering points in each of the multiple range-velocity distribution maps are extracted, and the intensity and weight of the scattering points corresponding to the coordinates are calculated. Based on the intensity and weight of the scattering points corresponding to the coordinates of each target scattering point, the weighted mean square residual, velocity weighted standard deviation, and distance weighted standard deviation corresponding to each range-velocity distribution map are calculated, and the corresponding complementary contrast factor is calculated through the velocity weighted standard deviation and the distance weighted standard deviation.

[0053] Those skilled in the art will understand that rigid targets of ships and passive jamming decoys (such as corner reflector arrays and chaff clouds) exhibit rigid / non-rigid differences in range-velocity distribution maps (RV Maps). Therefore, embodiments of this application can extract different features corresponding to the range-velocity distribution map based on this difference, including weighted mean square residuals and complementary contrast factors.

[0054] In this embodiment, the mean-weighted residual (MWR) can be calculated and used as one of the features of the range-velocity distribution map to highlight the consistency of motion of the scattering points of the rigid body target. In actual implementation, this embodiment can perform velocity-distance linear regression modeling on all scattering points in the RV Map, and calculate the mean-square value of the deviation between the actual velocity and the theoretical rigid body trajectory based on the weighted scattering intensity.

[0055] It is understandable that, due to the rigid structure of the ship as a whole, the motion of its scattering points exhibits strong spatial correlation, that is, there is a strong linear relationship between the velocity and distance of the scattering point, and the MWR value is small; while the scattering points of false targets such as corner reflector arrays move independently, and the relationship between the velocity and distance of the scattering point is more random, resulting in a significantly increased MWR.

[0056] Secondly, embodiments of this application may also design a complementary contrast factor F and use it as another feature of the distance-velocity distribution map to highlight the rigid body-specific pattern and reflect the distribution complementarity of the distance-velocity distribution map.

[0057] As one possible approach, embodiments of this application can calculate the normalized weighted standard deviations of the distance dimension and the velocity dimension separately, and calculate the complementary contrast factor F to analyze the complementary relationship between the dispersion of the two dimensions: due to rigid body constraints, the scattering point group (such as the ship hull) with a concentrated distribution in the distance dimension is dispersed in the velocity dimension, and vice versa, forming a complementary distribution of "one clustered and the other dispersed", resulting in a larger F value; the scattering points of the false target are in a disordered diffusion state, and the distribution in the two dimensions can be simultaneously dispersed or clustered, resulting in a smaller F value.

[0058] Therefore, the embodiments of this application combine the distribution characteristics of the distance-velocity distribution map and the rigid / non-rigid characteristics of the target for feature design and extraction, thereby effectively distinguishing ships from false targets, improving the accuracy of target identification, and providing reliable technical support for ship detection and anti-interference.

[0059] Optionally, in one embodiment of this application, based on the scattering point intensity and intensity weight corresponding to the coordinates of each target scattering point, the weighted mean square residual, velocity-weighted standard deviation, and distance-weighted standard deviation corresponding to each range-velocity distribution map are calculated, and the corresponding complementary contrast factor is calculated using the velocity-weighted standard deviation and distance-weighted standard deviation. This includes: performing a linear regression fitting operation on the scattering point coordinates on each range-velocity distribution map to obtain the corresponding linear regression model; determining the intensity weight corresponding to the scattering point intensity of each target scattering point coordinate, and calculating the weighted mean square residual corresponding to the scattering point coordinate based on the intensity weight and the linear regression model.

[0060] It should be noted that embodiments of this application can design two effective features to distinguish between rigid / non-rigid features such as ship targets and corner reflector array decoys on the target RV Map. For example... Figure 3 As shown, the specific feature extraction process is as follows: First, embodiments of this application can extract the coordinates of scattering points on each RV Map of the target. And calculate the weights corresponding to the scattering point intensities. ,in, Indicates the number of the scattering point.

[0061] In actual implementation, the following two features are designed and extracted for each RV Map data: (1) Feature 1: Weighted Mean Square Residual (MWR) In this embodiment, the coordinates of the scattering points on each RV Map can first be determined. A linear regression model was obtained by fitting the data. , in, These are the parameters used for linear regression fitting calculations.

[0062] Secondly, the embodiments of this application can calculate the residual sum of squares corresponding to the linear regression model of the scattering point coordinates. Considering that the contribution of different intensities of the scattering points to the features is not the same, the embodiments of this application need to weight the scattering points according to their corresponding intensities during the calculation process to obtain the weighted residual sum of squares (WRSS). The mathematical expression of the weighted residual sum of squares is as follows:

[0063] in, This represents the weighted sum of squared residuals; This indicates the number of scattering points on each distance-velocity distribution map; Indicates intensity weight; This represents the predicted value from the linear regression model; This represents the velocity value corresponding to the coordinates of the scattering point.

[0064] Considering the varying number of scattering points on each RV Map, this embodiment of the application can use a weighted sum of squared residuals. Divide by the number of scattering points to obtain the mean square weighted residual MWR, as shown in the following formula: .

[0065] (2) Feature 2: Complementary contrast factor F Based on the distribution of scattering point coordinates, the weighted standard deviations of the velocity and distance dimensions can be calculated in this embodiment, as shown in the following formula:

[0066]

[0067] Since the velocity and distance dimensions of the scattering point coordinates are not of the same order of magnitude, this embodiment of the application needs to align the order of magnitude of the data in the two dimensions of the scattering point coordinates before calculating the weighted standard deviation.

[0068] Based on the range-velocity distribution characteristics of scattering points on rigid / non-rigid targets, the range and velocity distributions of scattering points on rigid targets exhibit a complementary trend; that is, when the range dimension is more dispersed / concentrated, the velocity dimension is more concentrated / dispersed. Therefore, the complementary contrast factor can be calculated in the embodiments of this application. F This is the second feature of the RV Map.

[0069] Therefore, by combining intensity weights with linear regression models to calculate features, the embodiments of this application can more accurately characterize target features, improve target discrimination, and provide a reliable basis for target identification.

[0070] Optionally, in one embodiment of this application, the mathematical expression for the complementarity contrast factor is:

[0071] in, Indicates the complementary contrast factor; Indicates the distance from the weighted standard deviation; This represents the speed-weighted standard deviation; To prevent the denominator from being 0 for smaller values.

[0072] In the embodiments of this application, the complementarity contrast factor is as follows:

[0073] in, A very small number is set to prevent the denominator from being zero; generally speaking, ; and Same age, same time F When the value is small, the two dimensions show a "non-complementary" distribution, which can be identified as a non-rigid body target; while when the two dimensions show a "complementary" distribution, F If the value is relatively large, it can be used to identify it as a rigid target.

[0074] Therefore, the embodiments of this application can effectively distinguish between rigid and non-rigid targets through complementary contrast factors, thereby improving the accuracy and efficiency of target identification.

[0075] In step S103, feature vectors for all distance-velocity distribution maps are constructed based on weighted mean square residuals, complementary contrast factors, velocity-weighted standard deviations, and distance-weighted standard deviations. The feature vectors are then input into a support vector machine to obtain the corresponding rigid body identification model. The rigid body identification model is then used to identify the target to be tested, thereby obtaining the identification result corresponding to the target to be tested.

[0076] Subsequently, embodiments of this application can be based on two characteristics: weighted mean square residuals and complementary contrast factors, combined with two-dimensional dispersion. Together they form a four-dimensional discrimination vector (i.e., feature vector), which is then input into a pre-trained support vector machine network to obtain the identification result corresponding to the target to be tested.

[0077] It should be noted that this support vector machine-like network can construct a high-dimensional classification hyperplane through kernel function mapping, accurately segmenting rigid and non-rigid targets in the feature space. In actual deployment, the radar system can complete the entire process of "echo acquisition - RV Map generation - feature extraction - model classification" within a single scan cycle, thereby outputting the identification conclusion of ships / false targets.

[0078] Optionally, in one embodiment of this application, a rigid body identification model is used to identify the target to be tested in order to obtain the identification result corresponding to the target to be tested, including: determining whether the target to be tested is a rigid body based on the identification result; when the target type is a rigid body, determining that the target to be tested is a ship target; when the target type is a non-rigid body, determining that the target to be tested is a non-ship false target.

[0079] In actual implementation, the embodiments of this application can also include the weighted standard deviations of the two dimensions calculated in the RVMap feature vectors, thus including all RVMap feature vectors. The rigid body identification model (i.e., classification identification model) is obtained by inputting the training support vector machine. The target category is identified and determined using this model. Based on the identification result, it is determined whether the target category is a rigid body. If it is a rigid body, the target is identified as a ship target. If it is not a rigid body, the target is identified as a non-ship false target such as a corner reflector array.

[0080] The execution logic of the ship target identification method based on frequency-agile radar of this application will be explained below with reference to the accompanying drawings.

[0081] Figure 4 This is a schematic diagram illustrating the execution logic of the ship target identification method based on frequency-agile radar proposed in this application. Figure 4As shown, the execution process of the ship target identification method based on frequency-agile radar in this application is as follows: S401: Transmits pulse signals to the target under test via frequency-agile radar to obtain the corresponding echo signal from the target under test; S402: Perform a series of signal processing operations on the echo signal spectrum to generate the corresponding distance-velocity distribution map; S403: Extract different features corresponding to the distance-velocity distribution map, including weighted mean square residuals and complementary contrast factors; S404: Construct feature vectors based on weighted mean square residuals and complementary contrast factors, and input them into a support vector machine to obtain a classification and discrimination model; S405: Use the obtained classification and identification model to identify and identify the target to obtain the corresponding identification result.

[0082] In summary, this application embodiment, based on frequency-agile radar signals and processing, simultaneously obtains high range resolution determined by a large bandwidth B and high velocity resolution determined by a long coherence time T within a set of transmitted pulses. It utilizes frequency-agile radar to acquire the target's range-velocity distribution map and, combined with the rigid / non-rigid attribute differences between ship targets and passive jamming decoys (corner reflector arrays, chaff clouds, etc.), extracts multi-dimensional micro-motion features. Therefore, this application embodiment effectively overcomes the problems of low feature dimensionality and sensitivity to observation angle in HRRP, while reducing dependence on long-term observation and lowering computational complexity. Finally, the extracted multi-dimensional features are input into a support vector machine to obtain a target discrimination model, achieving efficient discrimination between ship targets and passive jamming targets, and improving the accuracy and generalization performance of target recognition.

[0083] According to the ship target identification method based on frequency-agile radar proposed in this application, a pulse signal is emitted to the target by a preset frequency-agile radar to obtain the corresponding echo signal. The echo signal is then subjected to pulse compression, Doppler compensation, and range migration compensation. An inverse Fourier transform is performed on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain a high-resolution range image of the target. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. These high-resolution range images are stitched together according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target. The scattering point coordinates of multiple target scattering points in each range-velocity distribution map are extracted, and the scattering... The scattering point intensity and intensity weight corresponding to the point coordinates are used to calculate the weighted mean square residual, velocity weighted standard deviation, and distance weighted standard deviation for each range-velocity distribution map based on the scattering point intensity and intensity weight corresponding to the scattering point coordinates of each target. The corresponding complementary contrast factor is calculated through the velocity weighted standard deviation and the distance weighted standard deviation. Based on the weighted mean square residual, complementary contrast factor, velocity weighted standard deviation, and distance weighted standard deviation, feature vectors of all range-velocity distribution maps are constructed. The feature vectors are input into a support vector machine to obtain the corresponding rigid body discrimination model. The rigid body discrimination model is used to identify the target to obtain the identification result of the target. This achieves efficient identification of ship targets and passive interference targets, improving the accuracy and generalization performance of the identification.

[0084] Secondly, with reference to the accompanying drawings, a ship target identification device based on frequency-agile radar according to an embodiment of this application is described.

[0085] Figure 5 This is a block diagram of a ship target identification device based on frequency-agile radar according to an embodiment of this application.

[0086] like Figure 5 As shown, the ship target identification device 10 based on frequency agile radar includes: a spectrum processing module 100, a feature extraction module 200, and a target identification module 300.

[0087] The spectrum processing module 100 is used to transmit pulse signals to the target under test through a preset frequency-agile radar to obtain the corresponding echo signal of the target under test. The echo signal is subjected to pulse compression, Doppler compensation and range migration compensation. The signal spectrum obtained after coherent accumulation and interpolation is subjected to inverse Fourier transform to obtain a high-resolution range image of the target under test. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. The series of corresponding high-resolution range images are stitched together according to the preset velocity dimension order to obtain multiple range and velocity distribution maps of the target under test.

[0088] The feature extraction module 200 is used to extract the scattering point coordinates of multiple target scattering points in each of the multiple range-velocity distribution maps, and calculate the scattering point intensity and intensity weight corresponding to the scattering point coordinates. Based on the scattering point intensity and intensity weight corresponding to each target scattering point coordinate, the module calculates the weighted mean square residual, velocity weighted standard deviation and distance weighted standard deviation corresponding to each range-velocity distribution map, and calculates the corresponding complementary contrast factor through the velocity weighted standard deviation and distance weighted standard deviation.

[0089] The target recognition module 300 is used to construct feature vectors for all distance and velocity distribution maps based on weighted mean square residuals, complementary contrast factors, velocity weighted standard deviations, and distance weighted standard deviations. The feature vectors are then input into a support vector machine to obtain the corresponding rigid body identification model. The rigid body identification model is used to identify the target to be tested, so as to obtain the identification result corresponding to the target to be tested.

[0090] Optionally, in one embodiment of this application, the spectrum processing module 100 includes: an equivalent unit, a demodulation unit, a processing unit, and a splicing unit.

[0091] The equivalent unit is used to transmit pulse signals to the target under test through the frequency agile radar and to convert the target under test into multiple corresponding target scattering points, so as to obtain the echo signal corresponding to each target scattering point based on the preset "stop-jump model".

[0092] The demodulation unit is used to perform linear superposition of the echo signals corresponding to each target scattering point, and demodulate the linearly superimposed echo signals to the baseband to obtain the echo signal spectrum corresponding to the target under test.

[0093] The processing unit is used to perform frequency domain matching processing on the echo signal spectrum, and to perform velocity Doppler compensation and range migration compensation operations on the frequency domain matching processed echo signal spectrum. It also performs coherent accumulation operation on the range migration compensated spectrum and frequency domain interpolation stitching on the coherently accumulated spectrum to unify the spectrum to the same frequency axis. Finally, it performs inverse Fourier transform on the spectrum unified to the same frequency axis to generate a high-resolution range image corresponding to the target under test.

[0094] The stitching unit is used to perform signal processing operations on the echo signal based on different preset velocity values ​​to generate multiple corresponding high-resolution range images, and stitch the multiple high-resolution range images according to the set velocity grid dimension order to obtain the range velocity distribution map of the target under test in each coherent processing interval of the frequency agile radar.

[0095] Optionally, in one embodiment of this application, the feature extraction module 200 includes a fitting unit and a calculation unit.

[0096] The fitting unit is used to perform linear regression fitting on the coordinates of the scattering points on each distance-velocity distribution map to obtain the corresponding linear regression model.

[0097] The calculation unit is used to determine the intensity weight corresponding to the scattering point intensity of each target scattering point coordinate, and calculate the weighted mean square residual corresponding to the scattering point coordinate based on the intensity weight and the linear regression model.

[0098] Optionally, in one embodiment of this application, the target recognition module 300 includes: a judgment unit, a first judgment unit, and a second judgment unit.

[0099] The judgment unit is used to determine whether the target to be tested is a rigid body based on the identification results.

[0100] The first determination unit is used to determine that the target to be tested is a ship target when the target type is a rigid body.

[0101] The second determination unit is used to determine that the target to be tested is a non-ship false target when the target type is a non-rigid body.

[0102] Optionally, in one embodiment of this application, the mathematical expression for the complementarity contrast factor is:

[0103] in, Indicates the complementary contrast factor; Indicates the distance from the weighted standard deviation; This represents the speed-weighted standard deviation; This is a minimal constant to prevent the denominator from being 0.

[0104] It should be noted that the foregoing explanation of the ship target identification method based on frequency-agile radar also applies to the ship target identification device based on frequency-agile radar in this embodiment, and will not be repeated here.

[0105] The ship target identification device based on frequency-agile radar proposed in this application includes a spectrum processing module 100, which is used to transmit pulse signals to the target under test through a preset frequency-agile radar to obtain the echo signal corresponding to the target under test. The echo signal is subjected to pulse compression, Doppler compensation, and range migration compensation. The signal spectrum obtained after coherent accumulation and interpolation is subjected to inverse Fourier transform to obtain a high-resolution range image of the target under test. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. The series of corresponding high-resolution range images are stitched together according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target under test. A feature extraction module 200 is used to extract the scattering coordinates of multiple target scattering points from each range-velocity distribution map in the multiple range-velocity distribution maps. The system calculates the scattering point intensity and intensity weight corresponding to the scattering point coordinates of each target. Based on the scattering point intensity and intensity weight corresponding to the scattering point coordinates of each target, it calculates the weighted mean square residual, velocity weighted standard deviation, and distance weighted standard deviation corresponding to each range-velocity distribution map. It also calculates the corresponding complementary contrast factor through the velocity weighted standard deviation and distance weighted standard deviation. The target recognition module 300 is used to construct feature vectors for all range-velocity distribution maps based on the weighted mean square residual, complementary contrast factor, velocity weighted standard deviation, and distance weighted standard deviation. The feature vectors are then input into a support vector machine to obtain the corresponding rigid body discrimination model. The rigid body discrimination model is used to identify the target to be tested, thereby obtaining the identification result corresponding to the target to be tested. This achieves efficient identification of ship targets and passive interference targets, improving the accuracy and generalization performance of the identification.

[0106] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 601, the processor 602, and the computer program stored on the memory 601 and capable of running on the processor 602.

[0107] When the processor 602 executes the program, it implements the ship target identification method based on frequency-agile radar provided in the above embodiments.

[0108] Furthermore, electronic devices also include: Communication interface 603 is used for communication between memory 601 and processor 602.

[0109] The memory 601 is used to store computer programs that can run on the processor 602.

[0110] The memory 601 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0111] If the memory 601, processor 602, and communication interface 603 are implemented independently, then the communication interface 603, memory 601, and processor 602 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0112] Optionally, in a specific implementation, if the memory 601, processor 602, and communication interface 603 are integrated on a single chip, then the memory 601, processor 602, and communication interface 603 can communicate with each other through an internal interface.

[0113] The processor 602 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0114] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described ship target identification method based on frequency-agile radar.

[0115] This application also provides a computer program product, including a computer program, which, when executed, is used to implement the above-described ship target identification method based on frequency-agile radar.

[0116] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0117] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0118] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0119] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0120] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0121] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0122] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0123] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for ship target identification based on frequency-agile radar, characterized in that, Includes the following steps: A pulse signal is transmitted to the target under test by a preset frequency-agile radar to obtain the corresponding echo signal of the target under test. The echo signal is subjected to pulse compression, Doppler compensation and range migration compensation. The signal spectrum obtained after coherent accumulation and interpolation is subjected to inverse Fourier transform to obtain a high-resolution range image of the target under test. Different velocity grid points are set and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. The series of corresponding high-resolution range images are stitched together according to the preset velocity dimension order to obtain multiple range velocity distribution maps of the target under test. Extract the scattering point coordinates of multiple target scattering points from each of the multiple range-velocity distribution maps, and calculate the scattering point intensity and intensity weight corresponding to the scattering point coordinates. Based on the scattering point intensity and intensity weight corresponding to each target scattering point coordinate, calculate the weighted mean square residual, velocity weighted standard deviation, and distance weighted standard deviation corresponding to each range-velocity distribution map, and calculate the corresponding complementary contrast factor through the velocity weighted standard deviation and the distance weighted standard deviation. Based on the weighted mean square residual, the complementary contrast factor, the velocity-weighted standard deviation, and the distance-weighted standard deviation, feature vectors of all distance-velocity distribution maps are constructed, and the feature vectors are input into a support vector machine to obtain the corresponding rigid body identification model. The rigid body identification model is then used to identify the target to be tested, so as to obtain the identification result corresponding to the target to be tested. The mathematical expression for the complementary contrast factor is: in, This represents the complementary contrast factor; This represents the distance-weighted standard deviation; This represents the speed-weighted standard deviation; This represents the numerical stability constant.

2. The method according to claim 1, characterized in that, The process involves transmitting pulse signals to the target using a preset frequency-agile radar to obtain the corresponding echo signals. The echo signals are then subjected to pulse compression, Doppler compensation, and range migration compensation. An inverse Fourier transform is performed on the signal spectrum obtained after coherent accumulation and interpolation stitching to obtain a high-resolution range image of the target. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. These high-resolution range images are then stitched together according to a preset velocity dimension order to obtain multiple range-velocity distribution maps of the target, including: The frequency-agile radar transmits the pulse signal to the target under test and converts the target under test into multiple corresponding target scattering points. Based on the preset "stop-jump model", the echo signal corresponding to each target scattering point is obtained. The echo signals corresponding to each target scattering point are linearly superimposed, and the linearly superimposed echo signals are demodulated to the baseband to obtain the echo signal spectrum corresponding to the target under test. The echo signal spectrum is subjected to frequency domain matching processing, and velocity Doppler compensation and range migration compensation operations are performed on the frequency domain matching processed echo signal spectrum. Coherent accumulation operation is performed on the range migration compensated spectrum, and frequency domain interpolation stitching is performed on the coherently accumulated spectrum to unify the spectrum to the same frequency axis. Inverse Fourier transform is performed on the spectrum unified to the same frequency axis to generate a high-resolution range image corresponding to the target under test. Based on different preset velocity values, the echo signal is processed to generate multiple high-resolution range profiles. These multiple high-resolution range profiles are then stitched together according to the set velocity grid dimension order to obtain the range velocity distribution map of the target under test in each coherent processing interval of the frequency agile radar.

3. The method according to claim 1, characterized in that, Based on the scattering point intensity and intensity weight corresponding to the coordinates of each target scattering point, the weighted mean square residual, velocity-weighted standard deviation, and distance-weighted standard deviation corresponding to each distance-velocity distribution map are calculated. The corresponding complementary contrast factor is then calculated using the velocity-weighted standard deviation and the distance-weighted standard deviation, including: A linear regression fitting operation is performed on the coordinates of the scattering points on each distance-velocity distribution map to obtain the corresponding linear regression model; Determine the intensity weight corresponding to the scattering intensity of each target scattering point coordinate, and calculate the weighted mean square residual corresponding to the scattering point coordinate based on the intensity weight and the linear regression model.

4. The method according to claim 1, characterized in that, The step of using the rigid body identification model to identify the target to be tested, so as to obtain the identification result corresponding to the target to be tested, includes: Based on the identification results, it is determined whether the target to be tested is a rigid body; When the target type is rigid body, the target to be tested is determined to be a ship target; When the target type is non-rigid body, the target to be tested is determined to be a non-ship false target.

5. A ship target identification device based on frequency-agile radar, characterized in that, include: The spectrum processing module is used to transmit pulse signals to the target under test through a preset frequency-agile radar to obtain the echo signal corresponding to the target under test. The echo signal is subjected to pulse compression, Doppler compensation and range migration compensation. The signal spectrum obtained after coherent accumulation and interpolation stitching is subjected to inverse Fourier transform to obtain a high-resolution range image of the target under test. Different velocity grid points are set, and the signal processing operation is repeated to obtain a series of corresponding high-resolution range images. The series of corresponding high-resolution range images are stitched together according to the preset velocity dimension order to obtain multiple range velocity distribution maps of the target under test. The feature extraction module is used to extract the scattering point coordinates of multiple target scattering points in each of the multiple distance-velocity distribution maps, and calculate the scattering point intensity and intensity weight corresponding to the scattering point coordinates. Based on the scattering point intensity and intensity weight corresponding to each target scattering point coordinate, the module calculates the weighted mean square residual, velocity weighted standard deviation, and distance weighted standard deviation corresponding to each distance-velocity distribution map, and calculates the corresponding complementary contrast factor through the velocity weighted standard deviation and the distance weighted standard deviation. The target recognition module is used to construct feature vectors for all distance-velocity distribution maps based on the weighted mean square residual, the complementary contrast factor, the velocity-weighted standard deviation, and the distance-weighted standard deviation, and input the feature vectors into a support vector machine to obtain the corresponding rigid body identification model, and use the rigid body identification model to identify the target to be tested, so as to obtain the identification result corresponding to the target to be tested; The mathematical expression for the complementary contrast factor is: in, This represents the complementary contrast factor; This represents the distance-weighted standard deviation; This represents the speed-weighted standard deviation; This represents the numerical stability constant.

6. The apparatus according to claim 5, characterized in that, The spectrum processing module includes: The equivalent unit is used to transmit the pulse signal to the target under test through the frequency agile radar, and to convert the target under test into multiple corresponding target scattering points, so as to obtain the echo signal corresponding to each target scattering point based on the preset "stop-jump model". The demodulation unit is used to perform a linear superposition operation on the echo signals corresponding to each target scattering point, and demodulate the linearly superimposed echo signals to the baseband to obtain the echo signal spectrum corresponding to the target under test. The processing unit is used to perform frequency domain matching processing on the echo signal spectrum, and to perform velocity Doppler compensation and range migration compensation operations on the echo signal spectrum after frequency domain matching processing, to perform coherent accumulation operation on the spectrum after range migration compensation, and to perform frequency domain interpolation stitching on the spectrum after coherent accumulation to unify the spectrum to the same frequency axis, and to perform inverse Fourier transform on the spectrum unified to the same frequency axis to generate a high-resolution range image corresponding to the target under test; The stitching unit is used to perform signal processing operations on the echo signal based on preset different velocity values ​​to generate multiple corresponding high-resolution range images, and to stitch the multiple high-resolution range images according to the set velocity grid dimension order to obtain the range velocity distribution map of the target under test in each coherent processing interval of the frequency agile radar.

7. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the ship target identification method based on frequency-agile radar as described in any one of claims 1-4.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the ship target identification method based on frequency-agile radar as described in any one of claims 1-4.

9. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the ship target identification method based on frequency-agile radar as described in any one of claims 1-4.

Images