A modeling method for omnidirectional scattering point characteristics of ship targets

Abstract

The present invention relates to a method for modeling the omnidirectional scattering point characteristics of a ship target, belonging to the technical field of radar target characteristic modeling. The method solves the problems of insufficient modeling granularity and low fidelity of electromagnetic models in the prior art. The method comprises: conducting an omnidirectional data acquisition test on a ship target in an outdoor field to obtain measured high-speed echo data; processing each frame of the measured high-speed echo data separately to obtain corresponding range-Doppler data; extracting target characteristics from the target area in each frame of range-Doppler data based on a normalized basis function to obtain target characteristic range-Doppler data corresponding to each frame of range-Doppler data; performing inter-frame correlation of target scattering points based on each frame of target characteristic range-Doppler data to obtain an omnidirectional stable scattering point sequence of the ship target; and performing omnidirectional scattering point modeling based on the omnidirectional stable scattering point sequence of the ship target to obtain an omnidirectional scattering point model of the ship.

Description

Technical Field

[0001] The present invention relates to the technical field of radar target characteristic modeling, and in particular to a method for modeling the omnidirectional scattering point characteristics of a ship target. Background Art

[0002] Radar's target detection and recognition performance directly impacts the combat effectiveness of weapon systems. Improving radar's ability to identify ship targets in complex battlefield environments requires extensive ship target data. By accurately analyzing ship target characteristics and simulating their echo signals, we construct a model of the omnidirectional scattering characteristics of ship targets. This allows for scene reconstruction and data augmentation of high-speed ship echo data, fully tapping into the value of test data and providing massive data support for improving weapon combat effectiveness.

[0003] However, the existing measured data is costly and difficult to obtain, and the simulation of radar target and environmental characteristics is difficult; at the same time, electromagnetic models often have low realism due to insufficient modeling granularity, and there are few studies on characteristics based on measured data.

[0004] Therefore, how to design a modeling method for the omnidirectional scattering point characteristics of ship targets is an urgent problem that needs to be solved. Summary of the Invention

[0005] In view of the above analysis, an embodiment of the present invention aims to provide a method for modeling the omnidirectional scattering point characteristics of a ship target, so as to solve the problems of insufficient modeling granularity and low fidelity of electromagnetic models in the prior art.

[0006] The present invention discloses a method for modeling characteristics of omnidirectional scattering points of a ship target, comprising:

[0007] Conduct omnidirectional data acquisition tests on ship targets in the field to obtain measured high-speed echo data; process each frame of measured high-speed echo data separately to obtain corresponding range-Doppler data;

[0008] Target characteristics are extracted from the target area in each frame of range-Doppler data based on the normalized basis function to obtain target characteristic range-Doppler data corresponding to each frame of range-Doppler data;

[0009] Based on the target characteristic range-Doppler data of each frame, the omnidirectional stable scattering point sequence of the ship target is obtained;

[0010] Omnidirectional scattering point modeling is performed according to the omnidirectional stable scattering point sequence of the ship target to obtain the ship omnidirectional scattering point model.

[0011] On the basis of the above solution, the present invention also makes the following improvements:

[0012] Furthermore, the omnidirectional stable scattering point sequence of the ship target is obtained by performing the following operations:

[0013] Perform decoherence processing on the target characteristic range-Doppler data of each frame, and take the prt with the largest amplitude as the one-dimensional range image of the frame;

[0014] Determining a one-dimensional range image sequence for a vertical sideways frame interval and a one-dimensional range image sequence for a non-vertical sideways frame interval based on the one-dimensional range image of each frame and its corresponding sideways angle;

[0015] Target characteristics are correlated with the one-dimensional range images of each frame in the non-vertical sideways frame interval to obtain a stable scattering point sequence in the non-vertical sideways frame interval;

[0016] Based on the stable scattering point sequence in the non-vertical sideways frame interval, a geometric model of the hull scattering position is established; based on the geometric model of the hull scattering position and the motion state of each stable scattering point in the stable scattering point sequence in the vertical sideways frame interval, a stable scattering point sequence in the vertical sideways frame interval is obtained;

[0017] The stable scattering point sequences in the non-vertical and vertical sideways frame intervals are sorted according to the frame sequence number to obtain the omnidirectional stable scattering point sequence of the ship target.

[0018] Furthermore, the stable scattering point sequence in the non-vertical sideways frame interval is obtained by performing the following operations:

[0019] According to the preset number of associated segment frames, the temporally adjacent one-dimensional range images in the non-vertical side frame interval are segmented, and the one-dimensional range images of each frame in the segment are sequentially spliced ​​according to the frame sequence number to obtain a plurality of segment data;

[0020] Preprocess each fragment data to obtain corresponding stable scattering point fragment data;

[0021] Perform range image principal component analysis and cluster analysis on each stable scattering point segment data, and output the stable scattering point sequence and cluster center associated between frames in the corresponding segment;

[0022] According to the stable scattering point sequence and cluster center associated between frames in each segment, the segment correlation is performed to obtain the stable scattering point sequence in the non-vertical side frame interval.

[0023] Furthermore, the stable scattering point sequence in the vertical sideways frame interval is obtained by performing the following operations:

[0024] According to the geometric model of the ship scattering position, the speed and heading of the ship target, the stable scattering point information of each stable scattering point in each broadside angle within the vertical broadside frame interval is determined;

[0025] According to the obtained stable scattering point information of each stable scattering point in the vertical sideways frame interval, a stable scattering point sequence in the vertical sideways frame interval is obtained.

[0026] Furthermore, determining the stable scattering point information of each stable scattering point at each side angle within the vertical side frame interval includes:

[0027] According to the geometric model of the hull scattering position, the speed and heading of the ship target, the radial velocity of the stable scattering point in the beam direction is calculated, and the radial velocity is converted into the corresponding Doppler channel number;

[0028] The Doppler channel number is mapped to the target characteristic range-Doppler data of the corresponding frame to determine the range position of the stable scattering point; the target scattering point corresponding to the range position is located in the one-dimensional range image of the corresponding frame as the stable scattering point, and the stable scattering point information of the located stable scattering point is obtained.

[0029] Furthermore, in the omnidirectional stable scattering point sequence of the ship target, the omnidirectional stable scattering point sequence of the kth stable scattering point is recorded as S k , S k It is expressed in the following way: k1 ,S k2 ,...,S kr ,...,S kn}; k = 1, 2, ..., K, K is the total number of stable scattering points; where S kr represents the stable scattering point information of the kth stable scattering point in the rth frame of range-Doppler data, where n is the total number of frames.

[0030] Furthermore, the stable scattering point information includes: frame number, side angle, range image position, scattering point phase and scattering point amplitude.

[0031] Furthermore, omnidirectional scattering points are modeled by performing the following operations:

[0032] Based on the broadside angle, the omnidirectional stable scattering point sequence is interpolated to fit the broadside angle, range image position, scattering point phase and scattering point amplitude, and the frame number is adjusted to realize omnidirectional scattering point modeling.

[0033] Furthermore, target characteristic range-Doppler data corresponding to each frame of range-Doppler data is obtained by performing the following operations:

[0034] Select the target area for each frame of range-Doppler data and annotate the selected target area with target information;

[0035] Obtain all target scattering points in the range-Doppler data of the target area;

[0036] The characteristic data of all target scattering points are superimposed and normalized to the radar-target centroid distance to obtain the target characteristic range-Doppler data of the current frame.

[0037] Furthermore, the target characteristic range-Doppler data C of the rth frame r Expressed as:

[0038]

[0039] d r =rangegate+D r .fbl (2)

[0040] Among them, d s is the target feature normalized distance, d r represents the actual distance between the radar and the center of mass of the ship target corresponding to the range-Doppler data of the rth frame, rangegate represents the gate position, and D r represents the range image position of the center of mass of the ship target in the rth frame of range-Doppler data, fbl represents the range resolution of the radar, and n is the total number of frames.

[0041] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0042] The present invention proposes a method for modeling the characteristics of omnidirectional scattering points of ship targets, which has the following beneficial effects:

[0043] First, the ship target characteristics are extracted from the radar high-speed echo data through a feature extraction algorithm based on basis functions. The problem of scattering point overlap in the association is solved through normalization calculation and hull geometry analysis. An omnidirectional scattering point association method is designed. Finally, the stable scattering point sequence is modeled to realize the reconstruction of the omnidirectional scattering point characteristics of the ship target and the simulation of the echo signal.

[0044] Second, the ship target omnidirectional scattering point characteristic modeling method of the present invention can effectively enhance the modeling granularity and improve the realism, solving the problem of low realism of the electromagnetic model due to insufficient modeling granularity.

[0045] Third, engineering practice has proved that the ship target omnidirectional scattering point characteristic modeling method proposed in the present invention can obtain a full-angle characterization model of the ship's azimuth, realize scene reconstruction and data augmentation of ship high-speed echo data, and provide a large amount of data support for radar target detection and recognition algorithm training, ultimately improving the radar's ability to identify ship targets and the combat effectiveness of weapons in complex battlefield environments.

[0046] In the present invention, the above-mentioned technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of the present invention will be described in the following description, and some advantages will become apparent from the description or be learned through practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like parts throughout the drawings.

[0048] Figure 1 Flowchart of the method for modeling the omnidirectional scattering point characteristics of a ship target provided by an embodiment of the present invention;

[0049] Figure 2 Schematic diagram of the correspondence between the range image position of the kth stable scattering point and the range image position of the center of mass of the ship target in the current frame when the broadside angle is 0° in the example;

[0050] Figure 3 Schematic diagram of the correspondence between the range image position of the kth stable scattering point and the range image position of the center of mass of the ship target in the current frame when the broadside angle is 45° in the example;

[0051] Figure 4 Schematic diagram of the method for obtaining the radial velocity of the stable scattering point in the beam direction in the example.

[0052] Figure 5 Schematic diagram of the ship's omnidirectional scattering point model structure. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, and are not used to limit the scope of the present invention.

[0054] A specific embodiment of the present invention discloses a method for modeling the characteristics of omnidirectional scattering points of a ship target, the flow chart of which is as follows: Figure 1 As shown, the following steps are included:

[0055] Step S1: conducting an omnidirectional data acquisition test on a ship target in the field to obtain measured high-speed echo data; processing each frame of measured high-speed echo data separately to obtain corresponding range-Doppler data;

[0056] In the field test scenarios constructed in this embodiment, each field test scenario contains only one ship target. After the field test scenarios are constructed, radar is used to collect echo data from the field test scenarios as the corresponding field-measured high-speed echo data. In this embodiment, pulse compression and coherent accumulation processing are performed on the measured high-speed echo data from the field test scenarios to obtain range-Doppler data corresponding to the measured high-speed echo data.

[0057] During the omnidirectional data acquisition test, the ship target rotates in a circle, so the entire time sequence of the measured high-speed echo data obtained corresponds to a 0-360° broadside angle.

[0058] Step S2: extracting target characteristics of the target area in each frame of range-Doppler data based on the normalized basis function to obtain target characteristic range-Doppler data corresponding to each frame of range-Doppler data;

[0059] Specifically, step S2 includes:

[0060] Step S21: Select a target area for each frame of range-Doppler data, and annotate the selected target area with target information;

[0061] Based on the mapping relationship between the field test scene and each frame of range-Doppler data, the target area can be framed out from each frame of range-Doppler data. After the target area is framed out, the target area can be annotated with target information. In this embodiment, the annotated target information includes: target attribute information, target location information, and target scene information;

[0062] Target attribute information includes target type and hull number / model; for example, target type includes warship and civilian ship, and hull number / model includes 835, etc.

[0063] The target location information refers to the range Doppler area where the target area is located;

[0064] Target parameter information includes: side angle, heading, speed and ship geometry (such as length, width and height);

[0065] Target scene information refers to the meteorological information of the field test scene, such as weather, sea conditions, wind speed, wind direction, ocean currents, ocean direction, etc.

[0066] Since in this embodiment, omnidirectional data collection is required for a ship target in an outdoor test scenario, each omnidirectional data collection test involves only one ship target. Accordingly, only one target area can be selected from the range-Doppler data.

[0067] In particular, if the outdoor scene also contains non-cooperative targets, the non-cooperative targets in the range-Doppler data are removed and replaced with background clutter;

[0068] Since non-cooperative targets (i.e., non-predetermined ship-like targets) inevitably appear in field measurement data, such as unknown small fishing boats and drilling platforms, they are marked as non-cooperative targets during data annotation. When extracting ship target characteristics, these non-cooperative targets need to be removed. Here, background clutter replacement is used to remove non-cooperative targets. Specifically, after the target area is selected and the target information is annotated, the range-Doppler range of the ship target and other non-cooperative targets is determined based on the annotated information of all targets in the current frame's range-Doppler data. The non-cooperative targets in the range-Doppler data are then removed and replaced with background clutter.

[0069] Step S22: Acquire all target scattering points in the range-Doppler data of the target area; specifically, including:

[0070] Step S221: obtaining the scattering point with the largest amplitude in the range-Doppler data of the target area;

[0071] Specifically, in step S221, when the process is executed for the first time, the range-Doppler data of the target area is determined based on the target position information of the marked target area. When the process is not executed for the first time, the range-Doppler data of the target area refers to the range-Doppler data of the target area updated in step S222.

[0072] Step S222: Determine whether the amplitude of the scattering point is greater than or equal to the clutter amplitude threshold.

[0073] If yes, the scattering point is extracted as a target scattering point, and the product of the amplitude of the scattering point and the normalized basis function is used as the characteristic data of the target scattering point;

[0074] Update the total number of target scattering points and determine whether the total number of target scattering points is greater than or equal to the preset number of target scattering points.

[0075] If so, then end;

[0076] Otherwise, the target scattering point is removed from the range-Doppler data of the target area, and the range-Doppler data of the target area is updated, and the process jumps to step S221, i.e., the scattering point with the largest amplitude is searched again in the updated range-Doppler data of the target area;

[0077] Otherwise, end;

[0078] After step S22 is performed (ie, after the process ends), characteristic data of all target scattering points can be obtained.

[0079] In step S222, a corresponding clutter amplitude threshold is determined based on the clutter data in the current frame range-Doppler data and a preset clutter probability threshold. Specifically, the clutter amplitude threshold is determined in the following manner: the clutter amplitude threshold β is calculated from the cumulative probability distribution P of the clutter data Z in the current frame range-Doppler data. The relationship between the clutter probability threshold α and the clutter amplitude threshold β is expressed as α=P(Z≤β), that is, the clutter probability threshold α is the probability that the clutter Z is less than or equal to the clutter amplitude threshold β. Therefore, when the clutter probability threshold is predetermined, the clutter amplitude threshold can be determined based on the clutter data in the current frame range-Doppler data. α∈(0,1], the higher the limit on the number of extracted target scattering points, the higher the value of the corresponding clutter probability threshold. At the same time, the amplitude of the extracted scattering points must be greater than the clutter amplitude threshold at the α quantile.

[0080] It should be noted that the normalized basis function in step S222 is determined in the following manner:

[0081] During the omnidirectional test of the ship target in the field test scenario, the radar equipment parameters remain unchanged. Therefore, in this embodiment, the baseband transmission signal is determined according to the radar equipment parameters, and the baseband transmission signal is convolved with itself to obtain the corresponding basis function set; the basis function set is coherently accumulated and normalized to obtain the normalized basis function. Exemplarily, in this embodiment, the radar equipment parameters include information such as the radar working state, polarization mode, working frequency band, signal bandwidth and code type. Specifically, the baseband transmission signal st is convolved with itself to obtain the basis function set of each PRT in the current frame, where the basis function Base of the i-th PRT is recorded as i for:

[0082]

[0083] Where Window is the matched filter window function, i = 1, 2, ..., m, m is the number of pulse accumulation, and the current frame basis function set Base is as follows:

[0084] Base=fft([Base1; Base2; ...; Base m ]) (2)

[0085] By coherently accumulating and normalizing the basis function set, the normalized basis function NBase can be obtained.

[0086] In step S222, the process of obtaining characteristic data of the target scattering point based on the normalized basis function and removing the target scattering point from the range-Doppler data of the target area is described as follows:

[0087] The scattering points are extracted in descending order according to the amplitude of the target scattering points in the range-Doppler data. Assume that the amplitude of the j-th target scattering point in the target area is I j , the coordinates are (a, b), the normalized basis function NBase is matrix translated (the redundant and missing parts are spliced ​​end to end after the translation), the basis of the translation is to move the maximum amplitude point of the normalized basis function NBase to (a, b), and the moved basis function is recorded as NBase j , will I j *NBase j As the characteristic data of the jth target scattering point in the current target area, I is then filtered out from the target area. j *NBase j , in order to eliminate the target scattering point from the range-Doppler data of the target area.

[0088] Step S23: The characteristic data of all target scattering points are superimposed and normalized by radar-target centroid distance to obtain target characteristic distance-Doppler data of the current frame.

[0089] For example, the target characteristic range-Doppler data C of the rth frame r Expressed as:

[0090]

[0091] d r =rangegate+D r .fbl (4)

[0092] Among them, d s is the target feature normalized distance, d r represents the actual distance between the radar and the center of mass of the ship target corresponding to the range-Doppler data of the rth frame, rangegate represents the gate position, and D r Indicates the range image position of the center of mass of the ship target in the range-Doppler data of the rth frame, fbl represents the range resolution of the radar, and n is the total number of frames. r It includes the data information of all target scattering points that constitute the target characteristics of the frame data. It is a two-dimensional RD complex matrix, which contains the distance, Doppler position and phase information of all target scattering points that constitute the target characteristics.

[0093] Step S3: Based on the target characteristic range-Doppler data of each frame, inter-frame correlation of the target scattering points is performed to obtain an omnidirectional stable scattering point sequence of the ship target; specifically, the process includes:

[0094] Step S31: perform decoherence processing on the target characteristic range-Doppler data of each frame, and take the prt with the largest amplitude (corresponding to a row in the decoherence data) as the one-dimensional range image of the frame;

[0095] Step S32: Determine, based on the one-dimensional range images of each frame and its corresponding sideways angle, a one-dimensional range image sequence of a vertical sideways frame interval and a one-dimensional range image sequence of a non-vertical sideways frame interval; wherein the non-vertical sideways frame interval refers to a frame interval in other sideways except the vertical sideways frame interval.

[0096] Since the radial size of the ship is the smallest in the vertical broadside frame interval (around plus or minus 90 degrees), the range images of each target scattering point will overlap. Therefore, the one-dimensional range images of the vertical broadside frame interval and the non-vertical broadside frame interval can be divided by judging whether the one-dimensional range images corresponding to each broadside angle overlap. At the same time, to solve this problem, this embodiment implements two different correlation technology solutions for the target characteristic range-Doppler data in the non-vertical broadside frame interval and the vertical broadside frame interval respectively. Specifically,

[0097] Step S33: performing target characteristic correlation on the one-dimensional range images of each frame in the non-vertical sideways frame interval to obtain a stable scattering point sequence in the non-vertical sideways frame interval;

[0098] Step S331: dividing the temporally adjacent one-dimensional range images in the non-vertical side frame interval into segments according to a preset number of associated segment frames, and sequentially splicing the one-dimensional range images of each frame in the segment according to the frame sequence number to obtain a plurality of segment data;

[0099] Step S332: pre-process each segment data to obtain corresponding stable scattering point segment data;

[0100] Since the scattering point characteristics do not necessarily exist at all broadside angles, the extracted target characteristic range-Doppler data contains several target scattering points with discontinuous and unstable time series. Therefore, it is necessary to preprocess each data segment, including outlier removal and missing value interpolation, to solve the problem of discontinuous and unstable time series of target scattering point data.

[0101] Outlier removal: Using the existing MAD algorithm, the target scattering point is judged as an outlier by determining whether the deviation between the range image of each target scattering point and the median value of the range images of all target scattering points is within a reasonable range. By executing this process, unstable target scattering points can be removed and the remaining target scattering points can be regarded as stable scattering points.

[0102] Missing value interpolation: Using the filling method of regression analysis, missing values ​​are calculated by fitting the regression mathematical model, that is, missing stable scattering points are filled.

[0103] After the preprocessing is completed, the stable scattering point segment data can be obtained. This application first performs intra-segment inter-frame correlation, and then performs inter-segment correlation to solve the problem that the omnidirectional data scattering points have a large span and are difficult to distinguish, effectively reducing the correlation difficulty and improving the accuracy of the correlation results. The following is the intra-segment inter-frame correlation and inter-segment correlation in turn:

[0104] Step S333: performing range image principal component analysis and cluster analysis on each stable scattering point segment data, and outputting the stable scattering point sequence and cluster centers associated between frames within the corresponding segment;

[0105] For each stable scattering point segment, first, principal component analysis (PCA) is performed on the range images of all stable scattering points within the segment to maximize their discrimination within the range image. Then, K-means clustering is performed on the range image using the PCA results, with the number of clustering categories equal to the number of stable scattering points. This results in a stable scattering point subsequence and its cluster center associated with each stable scattering point within the current segment. The stable scattering point subsequence records the stable scattering point information for each stable scattering point in each frame. This stable scattering point information includes: frame number, sideways angle, range image position, scattering point phase, and scattering point amplitude. The range image position, scattering point phase, and scattering point amplitude are obtained from the corresponding one-dimensional range image. The frame number refers to the frame data corresponding to the measured high-speed echo data collected in the field test scenario. The stable scattering point subsequences associated with each stable scattering point within the current segment are summarized to form a stable scattering point sequence associated with each stable scattering point within the corresponding segment.

[0106] Step S334: performing inter-segment correlation based on the inter-frame correlated stable scattering point sequences and cluster centers within each segment to obtain the stable scattering point sequences in the non-vertical sideways frame interval.

[0107] After all inter-frame correlations within each segment are completed, for adjacent stable scattering point segment data, the cluster centers of each stable scattering point segment data are predicted based on the ship's geometry and the stable scattering point positional relationships. Based on the distance relationship between the predicted cluster centers of each segment data and the cluster centers obtained by cluster analysis (step S333), the corresponding relationships between the same stable scattering point in each stable scattering point segment data are determined. Based on this corresponding relationship, inter-segment correlation is performed to obtain a stable scattering point sequence for the non-perpendicular sideways frame interval. Specifically, the inter-segment correlation method is as follows:

[0108] For the same target scattering point, the stable scattering point subsequences in different stable scattering point segments are spliced ​​according to the frame sequence number to form the stable scattering point sequence of the stable scattering point in the non-vertical sideways frame interval; the stable scattering point sequences of all stable scattering points in the non-vertical sideways frame interval are summarized to form the final stable scattering point sequence in the non-vertical sideways frame interval.

[0109] For target characteristic data association in the vertical broadside frame interval, since the radial size of the ship is the smallest in the vertical broadside frame interval, the range images of each target scattering point overlap. Therefore, it is necessary to combine the frequency domain information to associate the stable scattering points. The specific steps are as follows:

[0110] Step S34: Based on the stable scattering point sequence in the non-vertical sideways frame interval, a geometric model of the hull scattering position is established; based on the geometric model of the hull scattering position and the motion state of each stable scattering point in the stable scattering point sequence in the vertical sideways frame interval, a stable scattering point sequence in the vertical sideways frame interval is obtained; specifically,

[0111] Step S341: Establish a geometric model of the hull scattering position based on the stable scattering point sequence in the non-vertical sideways frame interval and the hull geometry; specifically,

[0112] Step S3411: establishing a hull geometric model based on the hull geometric structure;

[0113] Step S3412: For each stable scattering point, select several frames of stable scattering point information from the stable scattering point sequence in the non-vertical sideways frame interval, and determine the physical position of the stable scattering point in the established hull geometric model;

[0114] In step S3412, for each stable scattering point, each frame of stable scattering point information is selected, and the linear position of the stable scattering point at the current broadside angle is determined based on the corresponding relationship between the broadside angle and the range image position corresponding to the frame and the range image position of the center of mass of the ship target in the current frame. Based on the linear positions of the stable scattering point at multiple broadside angles, the physical position of the stable scattering point in the ship hull geometric model can be determined.

[0115] For example, for the kth stable scattering point, it is assumed that the stable scattering point sequence in the non-vertical sideways frame interval contains stable scattering point information with sideways angles of 0° and 45°; when the sideways angle is 0°, as Figure 2 As shown in , according to the correspondence between the range image position of the kth stable scattering point and the range image position of the center of mass of the ship target in the current frame, the radial position of the stable scattering point can be obtained. It can be seen that the kth scattering point is on a straight line in the hull that is perpendicular to the beam direction at this time. When the broadside angle is 45°, as Figure 3 As shown, it can be seen that the kth stable scattering point is located on another straight line in the hull that is perpendicular to the beam direction at this time. The intersection of these two lines is the physical location of the stable scattering point in the hull geometric model. More precisely, stable scattering point information corresponding to more broadside angles can be selected to more accurately determine the physical location of the stable scattering point in the hull geometric model. Similarly, the physical locations of all stable scattering points in the hull geometric model can be determined using this method.

[0116] Step S3413: Establishing a hull scattering position geometric model according to the physical position of each stable scattering point in the hull geometric model.

[0117] Step S342: determining the stable scattering point information of each stable scattering point in each broadside angle within the vertical broadside frame interval according to the ship scattering position geometric model, the speed and heading of the ship target;

[0118] Specifically, for each stable scattering point at each broadside angle within the vertical broadside frame interval, the following operations are performed:

[0119] Step S3421: Calculate the radial velocity of the stable scattering point in the beam direction based on the geometric model of the ship scattering position, the speed and heading of the ship target, and convert the radial velocity into the corresponding Doppler channel number;

[0120] Example: Calculate the stable scattering points (such as Figure 4 The Doppler channel number corresponding to the kth stable scattering point is shown in the figure (using the kth stable scattering point as an example): After obtaining the ship's motion state (speed and heading) for that frame, the velocity vector is projected onto the beam direction based on the ship's scattering position geometry model to obtain the radial velocity of the stable scattering point in the beam direction. The radial velocity is then converted to the corresponding Doppler channel number l according to the following formula:

[0121]

[0122] Where v is the radial velocity, PRT is the pulse repetition period, N is the Doppler dimension, and λ is the carrier wavelength.

[0123] Step S3422: The Doppler channel number is mapped to the target characteristic range-Doppler data of the corresponding frame to determine the range position of the stable scattering point; the target scattering point corresponding to the range position is located in the one-dimensional range image of the corresponding frame as the stable scattering point, and stable scattering point information of the located stable scattering point is obtained.

[0124] This process can solve the overlapping problem of the range images of the stable scattering point positions in the vertical side frame interval.

[0125] Step S343: Obtain a stable scattering point sequence in the vertical sideways frame interval according to the obtained stable scattering point information of each stable scattering point in the vertical sideways frame interval.

[0126] In the vertical sideways frame interval, for the same stable scattering point, the stable scattering point information is spliced ​​according to the frame sequence number to form a stable scattering point sequence for the stable scattering point in the vertical sideways frame interval; and the stable scattering point sequences of all stable scattering points in the vertical sideways frame interval are summarized to form a final stable scattering point sequence for the vertical sideways frame interval.

[0127] Step S35: sorting the stable scattering point sequences in the non-vertical and vertical sideways frame intervals according to the frame sequence numbers to obtain the omnidirectional stable scattering point sequence of the ship target.

[0128] In the omnidirectional stable scattering point sequence of the ship target, the omnidirectional stable scattering point sequence of the kth stable scattering point is recorded as S k , S k It can be expressed as follows: k1 ,S k2 ,...,S kr ,...,S kn}; k = 1, 2, ..., K, K is the total number of stable scattering points; where S kr Represents the stable scattering point information of the kth stable scattering point in the rth frame of range-Doppler data.

[0129] Step S4: performing omnidirectional scattering point modeling according to the omnidirectional stable scattering point sequence of the ship target to obtain a ship omnidirectional scattering point model.

[0130] In the omnidirectional stable scattering point sequence of a ship target, the deviation of the sideways angles corresponding to two adjacent frames may be relatively large. In this case, it is necessary to interpolate the omnidirectional stable scattering point sequence based on the sideways angle, fit the sideways angle, range image position, scattering point phase and scattering point amplitude, and adjust the frame number to achieve 360-degree omnidirectional scattering point modeling.

[0131] For example, the grid search method is used to select the corresponding fitting function to obtain the mathematical model of each scattering point of the ship target, and the omnidirectional scattering point model is performed to obtain the full-angle characterization model of the ship's azimuth, that is, the ship's omnidirectional scattering point model. The model structure diagram is shown in the figure below. Figure 5 Based on this model, target characteristics can be reconstructed according to the broadside angle, achieving incremental expansion of typical ship target sample data and scene reconstruction.

[0132] In summary, this paper proposes a method for modeling the omnidirectional scattering point characteristics of ship targets. This method extracts ship target characteristics from high-speed radar echo data using a basis function-based feature extraction algorithm. It also addresses the issue of scattering point overlap during correlation through normalization calculation and ship geometry analysis. An omnidirectional scattering point correlation method is designed. Finally, a stable scattering point sequence is modeled, enabling reconstruction of the omnidirectional scattering point characteristics of ship targets and echo signal simulation. Engineering practice has demonstrated that this method can generate a full-angle representation model of a ship's azimuth, enabling scene reconstruction and data augmentation of high-speed ship echo data. This method provides extensive data support for radar target detection and recognition algorithm training, ultimately improving radar's ship target recognition capabilities and weapon combat effectiveness in complex battlefield environments.

[0133] Those skilled in the art will appreciate that all or part of the process steps of the above-described embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, such as a magnetic disk, an optical disk, a read-only memory, or a random access memory.

[0134] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for modeling the characteristics of omnidirectional scattering points of a ship target, characterized by: include: Conduct omnidirectional data acquisition tests on ship targets in the field to obtain measured high-speed echo data; Process each frame of measured high-speed echo data separately to obtain the corresponding range-Doppler data; Target characteristics are extracted from the target area in each frame of range-Doppler data based on the normalized basis function to obtain target characteristic range-Doppler data corresponding to each frame of range-Doppler data; Based on the target characteristic range-Doppler data of each frame, the omnidirectional stable scattering point sequence of the ship target is obtained; Omnidirectional scattering point modeling is performed according to the omnidirectional stable scattering point sequence of the ship target to obtain the ship omnidirectional scattering point model; Obtain the omnidirectional stable scattering point sequence of the ship target by performing the following operations: Perform decoherence processing on the target characteristic range-Doppler data of each frame, and take the prt with the largest amplitude as the one-dimensional range image of the frame; Determining a one-dimensional range image sequence for a vertical sideways frame interval and a one-dimensional range image sequence for a non-vertical sideways frame interval based on the one-dimensional range image of each frame and its corresponding sideways angle; Target characteristics are correlated with the one-dimensional range images of each frame in the non-vertical sideways frame interval to obtain a stable scattering point sequence in the non-vertical sideways frame interval; Based on the stable scattering point sequence in the non-vertical sideways frame interval, a geometric model of the hull scattering position is established; based on the geometric model of the hull scattering position and the motion state of each stable scattering point in the stable scattering point sequence in the vertical sideways frame interval, a stable scattering point sequence in the vertical sideways frame interval is obtained; The stable scattering point sequences in the non-vertical and vertical sideways frame intervals are sorted according to the frame sequence number to obtain the omnidirectional stable scattering point sequence of the ship target.

2. The ship target omnidirectional scattering point characteristic modeling method according to claim 1, characterized in that: Obtain a stable scattering point sequence for a non-vertical broadside frame interval by performing the following operations: According to the preset number of associated segment frames, the temporally adjacent one-dimensional range images in the non-vertical side frame interval are segmented, and the one-dimensional range images of each frame in the segment are sequentially spliced ​​according to the frame sequence number to obtain a plurality of segment data; Preprocess each fragment data to obtain corresponding stable scattering point fragment data; Perform range image principal component analysis and cluster analysis on each stable scattering point segment data, and output the stable scattering point sequence and cluster center associated between frames in the corresponding segment; According to the stable scattering point sequence and cluster center associated between frames in each segment, the segment correlation is performed to obtain the stable scattering point sequence in the non-vertical side frame interval.

3. The ship target omnidirectional scattering point characteristic modeling method according to claim 1, characterized in that: Obtain a stable scattering point sequence for the vertical broadside frame interval by performing the following operations: According to the geometric model of the ship scattering position, the speed and heading of the ship target, the stable scattering point information of each stable scattering point in each broadside angle within the vertical broadside frame interval is determined; According to the obtained stable scattering point information of each stable scattering point in the vertical sideways frame interval, a stable scattering point sequence in the vertical sideways frame interval is obtained.

4. The ship target omnidirectional scattering point characteristic modeling method according to claim 3 is characterized in that: The determining of the stable scattering point information of each stable scattering point at each side angle within the vertical side frame interval includes: According to the geometric model of the hull scattering position, the speed and heading of the ship target, the radial velocity of the stable scattering point in the beam direction is calculated, and the radial velocity is converted into the corresponding Doppler channel number; The Doppler channel number is mapped to the target characteristic range-Doppler data of the corresponding frame to determine the range position of the stable scattering point; the target scattering point corresponding to the range position is located in the one-dimensional range image of the corresponding frame as the stable scattering point, and the stable scattering point information of the located stable scattering point is obtained.

5. The ship target omnidirectional scattering point characteristic modeling method according to any one of claims 1 to 4, characterized in that: In the omnidirectional stable scattering point sequence of the ship target, the omnidirectional stable scattering point sequence of the kth stable scattering point is recorded as S k , S k It is expressed in the following way: k1 ,S k2 ,...,S kr ,...,S kn }; k = 1, 2, ..., K, K is the total number of stable scattering points; where S kr represents the stable scattering point information of the kth stable scattering point in the rth frame of range-Doppler data, where n is the total number of frames.

6. The ship target omnidirectional scattering point characteristic modeling method according to claim 5, characterized in that: The stable scattering point information includes: frame number, side angle, range image position, scattering point phase and scattering point amplitude.

7. The ship target omnidirectional scattering point characteristic modeling method according to claim 6, characterized in that: Model an omnidirectional scattering point by doing the following: Based on the broadside angle, the omnidirectional stable scattering point sequence is interpolated to fit the broadside angle, range image position, scattering point phase and scattering point amplitude, and the frame number is adjusted to realize omnidirectional scattering point modeling.

8. The ship target omnidirectional scattering point characteristic modeling method according to claim 1, characterized in that: Obtain the target characteristic range-Doppler data corresponding to each frame of range-Doppler data by performing the following operations: Select the target area for each frame of range-Doppler data and annotate the selected target area with target information; Obtain all target scattering points in the range-Doppler data of the target area; The characteristic data of all target scattering points are superimposed and normalized to the radar-target centroid distance to obtain the target characteristic range-Doppler data of the current frame.

9. The ship target omnidirectional scattering point characteristic modeling method according to claim 1, characterized in that: The target characteristic range-Doppler data C of the rth frame r Expressed as: d r =rangegate+D r .fbl (2) Among them, d s is the target feature normalized distance, d r I represents the actual distance between the radar and the center of mass of the ship target corresponding to the range-Doppler data of the rth frame; j Represents the amplitude of the j-th target scattering point in the target area; the normalized basis function NBase is matrix translated. The basis of the translation is to move the point with the maximum amplitude of the normalized basis function NBase to the coordinate (a, b) of the j-th target scattering point. The moved basis function is recorded as NBase j ; rangegate represents the gate position, D r represents the range image position of the center of mass of the ship target in the rth frame of range-Doppler data, fbl represents the range resolution of the radar, and n is the total number of frames.

Images