A method and system for three-dimensional imaging of sparse data of a millimeter wave radar

Abstract

The application discloses a kind of millimeter wave radar sparse data three-dimensional imaging method and system, the method comprises: obtaining the three-dimensional echo data of sampling point;The three-dimensional echo data is decomposed in height direction section to obtain Nh layer two-dimensional echo data;Observation equation is constructed based on the two-dimensional echo data;Based on the observation equation, Nh two-dimensional image is obtained by first algorithm;The coherent superposition processing of signal peak value is carried out to each pixel in azimuth direction section to Nh two-dimensional image, to obtain the three-dimensional image of the target point.The application solves the problem of imaging technology index deterioration when the conventional imaging method is used for imaging sparse data.Under the premise of imaging index deterioration, the application of the application can maintain good imaging quality, and the application is not limited to imaging processing of sparse data.

Description

Technical Field

[0001] This application relates to the field of radar imaging technology, and in particular to a method and system for three-dimensional imaging of sparse data from millimeter-wave radar. Background Technology

[0002] Millimeter-wave radar imaging technology is an imaging technique that actively emits millimeter waves to acquire the scattering information of the target, thereby achieving three-dimensional imaging of the target. Millimeter-wave radar is a high-resolution imaging radar; millimeter waves can penetrate clothing without harming the human body, imaging dangerous and concealed targets hidden within the human body, thus improving public safety. Millimeter-wave radar is typically used in security inspection equipment, with an observation range within 1 meter. Conventionally acquired three-dimensional echo data of the target is characterized by large data volume, significant redundancy, and slow imaging speed.

[0003] In millimeter-wave radar 3D imaging geometry, a combination of real aperture and synthetic aperture is employed, along with mechanical scanning motion to achieve planar aperture synthesis, thereby realizing high-resolution 3D imaging of the target. In this imaging geometry, to achieve high-precision 3D imaging of the target, the processing of radar echo data must satisfy classical signal processing theory; that is, signal sampling must satisfy the Nyquist sampling theorem. Therefore, more antenna elements are required, and the amount of echo data obtained is enormous.

[0004] Conventional imaging methods are ineffective for sparse data. Current research on cylindrical aperture human body security imaging equipment proposes a frequency-domain sparse imaging method and a sparse array design. This method uses uniform sampling in the array direction (height) and Barker code random sparse sampling in the azimuth direction to reduce the amount of data required for imaging and system complexity, while also improving imaging speed. However, this method primarily focuses on interferometric processing and frequency-domain compressed sensing imaging of echo data randomly sampled in the mechanical scanning direction of the cylindrical aperture imaging model. It does not yet consider imaging dimensionally based on sparse echo data acquired from planar aperture imaging models to improve data acquisition efficiency and maintain imaging performance even with sparse data. Summary of the Invention

[0005] One objective of this application is to provide a method for three-dimensional imaging of sparse data from millimeter-wave radar, including:

[0006] Acquire three-dimensional echo data of sampling points, wherein the sampling points are used to scan the target point;

[0007] The three-dimensional echo data is decomposed in the height-oriented section to obtain two-dimensional echo data of the Nh layer;

[0008] An observation equation is constructed based on the two-dimensional echo data;

[0009] Based on the observation equation, Nh two-dimensional images are obtained using the first algorithm;

[0010] The signal peaks of each pixel in the Nh images along the azimuth cross section are coherently superimposed to obtain a three-dimensional image of the target point.

[0011] As an optional embodiment, acquiring the three-dimensional echo data of the sampling points includes:

[0012] Define a planar scanning 3D imaging model, where the coordinates of the target point are (xn, yn, zn) and the coordinates of the sampling point are (x', 0, z').

[0013] Based on the planar scanning 3D imaging model, the distance between the sampling point and the geometric center of the target point is calculated.

[0014] As an optional embodiment, constructing the observation equation based on the two-dimensional echo data includes:

[0015] Based on a preset step size, the preset region is divided into imaging region grids, and the division results are obtained;

[0016] Based on the partitioning results, a distance matrix is ​​constructed;

[0017] Based on the distance matrix, construct the observation matrix;

[0018] Based on the division results, an imaging region grid matrix is ​​constructed;

[0019] Based on the two-dimensional echo data, a two-dimensional echo matrix is ​​constructed;

[0020] The observation equation is constructed based on the distance matrix, the observation matrix, the imaging region raster matrix, and the two-dimensional echo matrix.

[0021] As an optional embodiment, based on a preset step size, a preset region is divided into an imaging region grid, and the division result is obtained, including:

[0022] The preset region in the distance direction is divided into imaging grids by a preset first step length to obtain multiple distance units;

[0023] The preset region in the azimuth direction is divided into an imaging grid by a preset second step length to obtain multiple azimuth units;

[0024] The segmentation result is obtained based on multiple distance units and multiple orientation units.

[0025] As an optional embodiment, the step of obtaining Nh two-dimensional images based on the observation equation using a first algorithm includes:

[0026] The scattering information column vector of the imaging region grid matrix is ​​calculated using the first formula, and the scattering information column vector is rearranged into a scattering information matrix to obtain a two-dimensional image.

[0027] The first count variable is processed, and the processed first count variable is compared with the first decomposition number, wherein the first count variable corresponds to the position of the two-dimensional image in the height section, and the first decomposition number is the decomposition number of the three-dimensional echo data;

[0028] Based on the first comparison result, determine whether to continue the step of calculating the scattering information column vector until Nh two-dimensional images are obtained.

[0029] As an optional embodiment, the step of determining whether to continue calculating the scattering information column vector based on the first comparison result until Nh two-dimensional images are obtained includes:

[0030] If the processed first count variable falls within the range of the first decomposition quantity, then continue to execute the step of calculating the scattering information column vector;

[0031] If the processed first count variable does not fall within the range of the first decomposition number, then stop the step of calculating the scattering information column vector to obtain Nh two-dimensional images corresponding to the first decomposition number.

[0032] As an optional embodiment, the coherent superposition of signal peaks of each pixel in the azimuth section of the Nh images of the two-dimensional image to obtain a three-dimensional image of the target point includes:

[0033] For each pixel in the azimuth section of each of the two-dimensional images, coherent superposition of signal peaks is performed;

[0034] Construct a corresponding matched filter matrix for each of the processed azimuth tangents;

[0035] Process the second count variable and compare the second count variable with the second decomposition quantity, wherein the second count variable corresponds to the position of the two-dimensional image in the azimuth section, and the second decomposition quantity is the number of azimuth grid cells;

[0036] Based on the second comparison result, it is determined whether to continue the step of constructing the matched filter matrix until a three-dimensional image of the target point is obtained.

[0037] As an optional embodiment, constructing the corresponding matched filter matrix for each processed azimuth tangent includes:

[0038] Construct the distance matrix from the sampling point to each of the azimuth tangents;

[0039] Process the distance matrix to obtain the matched filter matrix;

[0040] Multiply each pixel in the azimuth section with its corresponding matched filter matrix to obtain the matched signal;

[0041] The matching signal is subjected to distance compression processing;

[0042] The peak values ​​of the distance compression signals of each pixel after transformation are coherently superimposed to obtain the superposition result of the peak values ​​of the distance compression signals of all pixels in the azimuth section.

[0043] As an optional embodiment, the step of determining whether to continue executing the step of constructing a matched filter matrix based on the second comparison result until a three-dimensional image of the target point is obtained includes:

[0044] If the processed second count variable falls within the range of the second decomposition quantity, then continue with the step of constructing the matched filter matrix;

[0045] If the processed second count variable does not fall within the range of the second decomposition quantity, then the step of constructing the matched filter matrix is ​​stopped to obtain the three-dimensional image corresponding to the target point.

[0046] One of the objectives of this application is to provide a millimeter-wave radar sparse data three-dimensional imaging system, characterized in that it includes:

[0047] The first acquisition module is configured to acquire three-dimensional echo data of sampling points, wherein the sampling points are used to scan the target point;

[0048] The decomposition module is configured to decompose the three-dimensional echo data in the height-oriented tangent to obtain two-dimensional echo data of the Nh layer.

[0049] The construction module is configured to construct observation equations based on the two-dimensional echo data;

[0050] The second acquisition module is configured to acquire Nh two-dimensional images based on the observation equation and using the first algorithm.

[0051] The third acquisition module is configured to perform coherent superposition processing of signal peaks on each pixel in the azimuth section of the Nh images of the two-dimensional image to obtain a three-dimensional image of the target point.

[0052] The beneficial effects of the embodiments of this application are as follows:

[0053] This invention solves the problem of deterioration in imaging performance when conventional imaging methods are used to image sparse data. Applying this invention under conditions of deteriorated imaging performance can maintain good image quality. This invention is not limited to imaging processing of sparse data.

[0054] This invention establishes a frequency-domain compressed sensing measurement model for imaging processing of sparse data. It applies compressed sensing reconstruction technology to millimeter-wave radar three-dimensional imaging systems. By performing two-dimensional CS imaging of the echo signal in the range and azimuth directions, the amount of echo data is reduced to a certain extent, which can effectively alleviate the problem of large storage pressure caused by large data volume, improve imaging efficiency, and the frequency-domain CS imaging technology based on the echo signal can greatly suppress sidelobes, improve resolution and data acquisition efficiency. Attached Figure Description

[0055] Figure 1 This is a flowchart of a millimeter-wave radar sparse data three-dimensional imaging method according to an embodiment of this application;

[0056] Figure 2 This is a flowchart of the millimeter-wave radar sparse data three-dimensional imaging method S1 according to an embodiment of this application;

[0057] Figure 3 This is a flowchart of the millimeter-wave radar sparse data three-dimensional imaging method S3 according to an embodiment of this application;

[0058] Figure 4 This is a flowchart of S31 of the millimeter-wave radar sparse data three-dimensional imaging method according to an embodiment of this application;

[0059] Figure 5 This is a flowchart of the millimeter-wave radar sparse data three-dimensional imaging method S4 according to an embodiment of this application;

[0060] Figure 6 This is a flowchart of S43 of the millimeter-wave radar sparse data three-dimensional imaging method according to an embodiment of this application;

[0061] Figure 7 This is a flowchart of the millimeter-wave radar sparse data three-dimensional imaging method S5 according to an embodiment of this application;

[0062] Figure 8 This is a flowchart of S52 of the millimeter-wave radar sparse data three-dimensional imaging method according to an embodiment of this application;

[0063] Figure 9 This is a flowchart of S54 of the millimeter-wave radar sparse data three-dimensional imaging method according to an embodiment of this application;

[0064] Figure 10 This is a structural block diagram of a millimeter-wave radar sparse data three-dimensional imaging system according to an embodiment of this application;

[0065] Figure 11 This is a simplified structural diagram of the millimeter-wave radar sparse data three-dimensional imaging method S1 according to an embodiment of this application.

[0066] Figure label:

[0067] 100. First acquisition module; 200. Decomposition module; 300. Construction module; 400. Second acquisition module; 500. Third acquisition module. Detailed Implementation

[0068] Various embodiments and features of this application are described herein with reference to the accompanying drawings.

[0069] It should be understood that various modifications can be made to the embodiments described herein. Therefore, the above description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this application will be apparent to those skilled in the art.

[0070] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0071] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0072] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application.

[0073] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0074] Specific embodiments of this application are described thereafter with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in a variety of substantially any suitable detailed structures.

[0075] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.

[0076] When echo data is sparse, conventional imaging methods, such as the Range Migration Algorithm (RMA) or the Range Doppler Algorithm (RDA), cannot produce normal images, and the imaging quality metrics, such as the peak sidelobe ratio (PSLR) and integral sidelobe ratio (ISLR), will deteriorate significantly.

[0077] Therefore, one of the objectives of this application is to provide a method for three-dimensional imaging of sparse data from millimeter-wave radar, such as... Figure 1 As shown, it includes:

[0078] S1. Acquire the three-dimensional echo data of the sampling points, wherein the sampling points are used to scan the target point. Specifically, the three-dimensional echo data S(x',z',kω) of the millimeter-wave radar imaging system is known, wherein the echo data S(x',z',kω) is N a ×K×N h The matrix is ​​a three-dimensional complex matrix, where Na is the number of sampling points in the azimuth direction, Nh is the number of sampling points in the altitude direction, and K is the number of sampling points in the range direction.

[0079] As an optional embodiment, such as Figure 2 and Figure 11 As shown, acquiring the three-dimensional echo data of the sampling points includes:

[0080] S11. Determine the planar scanning 3D imaging model, where the coordinates of the target point are (xn, yn, zn), and the coordinates of the sampling point are (x', 0, z'). Specifically, define X'OZ' as the antenna scanning plane, the distance R between the geometric center of the target and the antenna element, the coordinates of any target point Pn in the imaging area are (xn, yn, zn), and the coordinates of any sampling point P in the planar aperture are (x', 0, z').

[0081] The planar scanning 3D imaging model obtains a high-resolution complex image of the target in the azimuth direction through a sparse array antenna structure. The sparse array antenna obtains a high-resolution complex image in the height direction through mechanical scanning. A high-resolution image in the range direction is achieved by transmitting and receiving millimeter-wave signals using each sparse array element. The target scattering coefficient is σn. The geometric properties of the planar aperture 3D imaging are as follows: Figure 11 As shown.

[0082] S12. Based on the planar scanning three-dimensional imaging model, calculate the distance between the sampling point and the geometric center of the target point.

[0083] Specifically, the three-dimensional echo data from the millimeter-wave radar is known:

[0084]

[0085] Wherein, the scattered echo signal s(x',z',kω) is N a' ×K×N h The matrix is ​​a three-dimensional complex matrix, where Na' is the number of sparse sampling points in the azimuth direction, Nh is the number of sampling points in the altitude direction, K is the number of sampling points in the range direction, x' represents the coordinates of the sampling points on the X' axis, z' represents the coordinates of the sampling points on the Z' axis, and kω represents the magnitude of the wavenumber vector corresponding to the stepped-frequency continuous wave signal. The relationship between kω and the instantaneous frequency f and the speed of light c is k ω =2πf / c, k ω ∈[k ωmin ,k ωmax ], kωmin and kωmax represent the magnitudes of the wavenumber vectors corresponding to the lowest and highest frequencies, respectively.

[0086] The distance between sampling point P and the geometric center of target Pn(xn,yn,zn) is:

[0087]

[0088] S2. Decompose the three-dimensional echo data in the height direction section to obtain the two-dimensional echo data of the Nh layer, which facilitates the subsequent two-dimensional imaging of all two-dimensional echo data in the azimuth-range direction, i.e., the height direction section.

[0089] Specifically, the m-th two-dimensional cross-sectional data is taken from the three-dimensional echo data along the azimuth-range two-dimensional plane, denoted as S. ikZm Where i represents the i-th sampling in the azimuth direction, and i∈[1,N] a '], k represents the k-th sampling in the distance direction, and k∈[1,K], Z m This indicates that the height is sampled for the mth time, and m∈[1,Nh], and the data acquisition operation starts when m=1.

[0090] S3. Construct observation equations based on the two-dimensional echo data. Specifically, construct observation equations for the selected m-th two-dimensional cross-sectional data.

[0091] As an optional embodiment, such as Figure 3 As shown, the construction of the observation equation based on the two-dimensional echo data includes:

[0092] S31. Divide the preset region into imaging region grids based on a preset step size, and obtain the division results. For example... Figure 4 As shown, it includes:

[0093] S311. Divide the preset region in the range direction into imaging grids using a preset first step length to obtain multiple range units. Specifically, divide the preset region in the imaging range direction [Y] into imaging grids. a ,Yb [With the first step length Y] c The imaging grid is divided into N grids. y One distance unit.

[0094] S312. Divide the preset region in the azimuth direction into an imaging grid using a preset second step length to obtain multiple azimuth units. Specifically, divide the preset region in the imaging azimuth direction [X... a ,X b With the second step length X c The imaging grid is divided into N grids. x One directional unit.

[0095] S313. Based on the multiple distance cells and the multiple azimuth cells, obtain the partitioning result for subsequent construction of the observation matrix. Specifically, the partitioning result is N = N x N y .

[0096] S32. Based on the partitioning results, construct a distance matrix. Specifically, the distance matrix is ​​defined as follows: Where N = N x N y , The distance matrix from the sampling points to the imaging region is shown in equation (3):

[0097]

[0098] in, This indicates the m-th data point in terms of altitude and the N-th data point in terms of azimuth. a' The k-th sampling point represents the distance from the sampling point to the nth target, where K represents the number of sampling points in the distance direction.

[0099] S33. Based on the distance matrix, construct the observation matrix. Specifically, for the constructed distance matrix... The observation matrix is ​​obtained through processing. As shown in equation (4), where N a' Z is the number of sampling points in the azimuth direction, K is the number of sampling points in the range direction, and Z is the number of sampling points in the range direction. Km This indicates that the data taken along the height direction is the m-th data point, and N = N. x N y This represents the total number of all grid cells in the imaging area.

[0100]

[0101] S34. Based on the division results, construct the imaging region raster matrix; construct the imaging region raster matrix. As shown in equation (5):

[0102]

[0103] in, This indicates that in the m-th data point, the imaging region's azimuth is represented by the N-th data point. x The nth grid, distance to the Nth grid y The scattering coefficients of the target in each grid cell are rearranged σ' in column vector form as follows: As shown in equation (6):

[0104]

[0105] S35. Based on the two-dimensional echo data, construct a two-dimensional echo matrix. Specifically, the two-dimensional echo matrix... As shown in equation (7):

[0106]

[0107] in, This represents the m-th two-dimensional echo matrix in the height direction. Indicates the direction to the Nth a' The next sample, the distance to the echo value of the Kth sample, will Rearranged in column vector form as follows As shown in equation (8):

[0108]

[0109] S36. Construct an observation equation based on the distance matrix, the observation matrix, the imaging region raster matrix, and the two-dimensional echo matrix. Specifically, the observation equation is shown in equation (9):

[0110]

[0111] in, For the observation matrix, The target scattering coefficient matrix,

[0112] Let m be the height-direction echo matrix.

[0113] S4. Based on the observation equation, obtain Nh two-dimensional images using the first algorithm;

[0114] As an optional embodiment, such as Figure 5 As shown, the acquisition of Nh two-dimensional images based on the observation equation using the first algorithm includes:

[0115] S41. Calculate the scattering information column vector of the imaging region grid matrix using the first formula, and rearrange the scattering information column vector into a scattering information matrix to obtain a two-dimensional image.

[0116] Specifically, the first formula is shown in equation (10):

[0117]

[0118] First, set the first counting variable to m, m = 1; calculate the height-oriented two-dimensional echo data of the m-th line (i.e., the preset position) according to formula (10). The scattering information column vector σ” is then rearranged into the form of the scattering information matrix σ', denoted as σ'. This is the first two-dimensional image.

[0119] S42. Process the first count variable and compare the processed first count variable with the first decomposition number, wherein the first count variable corresponds to the position of the two-dimensional image in the height-direction section, and the first decomposition number is the decomposition number of the three-dimensional echo data. Specifically, the first decomposition number is Nh.

[0120] S43. Based on the first comparison result, determine whether to continue the step of calculating the scattering information column vector until Nh two-dimensional images are obtained.

[0121] As an optional embodiment, such as Figure 6 As shown, the step of determining whether to continue calculating the scattering information column vector based on the first comparison result until Nh two-dimensional images are obtained includes:

[0122] S431. If the processed first count variable falls within the range of the first decomposition quantity, then continue to execute the step of calculating the scattering information column vector;

[0123] S432. If the processed first count variable does not fall within the range of the first decomposition number, then stop the step of calculating the scattering information column vector to obtain Nh two-dimensional images corresponding to the first decomposition number.

[0124] Specifically, let the first counting variable m+1, if m≤N h Continue with step S42; otherwise, stop execution, where N h This represents the number of sampling points along the height. After execution, N will be obtained. h Two-dimensional image, recorded N represents h A two-dimensional image, in which

[0125] S5. Perform coherent superposition of signal peaks on each pixel in the azimuth section of the Nh images of the two-dimensional image to obtain a three-dimensional image of the target point.

[0126] As an optional embodiment, such as Figure 7 As shown, the process of overlaying Nh two-dimensional images to obtain a three-dimensional image of the target point includes:

[0127] S51. Perform coherent superposition of signal peaks on each pixel in the azimuth section of each two-dimensional image.

[0128] Specifically, for N h For each pixel in a two-dimensional image I along the azimuth sectional plane (range-height two-dimensional plane), coherent superposition of signal peaks is performed. The p-th azimuth sectional plane is taken and denoted as... in Set the second counter variable to p, p = 1.

[0129] S52. Construct a corresponding matched filter matrix for each processed two-dimensional image. Specifically, constructing the azimuth sectional plane is equivalent to constructing a matched filter matrix for a two-dimensional plane in the range-height direction.

[0130] As an optional embodiment, such as Figure 8 As shown, constructing the corresponding matched filter matrix for each processed azimuth tangent includes:

[0131] S521. Construct the distance matrix from the sampling point to each of the azimuth tangents.

[0132] Specifically, construct sampling points to N h The distance matrix of the p-th azimuth tangent plane, i.e., the p-th range-elevation two-dimensional plane, in the two-dimensional image I is: Where N' = N y N h N' represents the number of pixels in the range-height two-dimensional plane, and K represents the number of radar range sampling points. y N represents the distance raster count, which is equivalent to the distance pixel count. h Represents the number of height-based sampling points, distance matrix As shown in equation (11):

[0133]

[0134] in, Indicates the height towards the Nth h The k-th sampling point is located at a distance from the sampling point to N. h The distance to the nth pixel in the p-th azimuth sectional plane of the two-dimensional image I, which is the p-th distance-height two-dimensional plane.

[0135] S522. Process the distance matrix to obtain the matched filter matrix. Specifically, construct the matched filter matrix for the p-th azimuth tangent, i.e., the p-th range-elevation two-dimensional plane, as follows: The distance matrix constructed in step S521 The process is performed to obtain the matched filter matrix. As shown in equation (12):

[0136]

[0137] S523. Multiply each pixel in the azimuth section with its corresponding matched filter matrix to obtain a matched signal. Specifically, multiply each pixel in the p-th azimuth section, i.e., the p-th range-height two-dimensional plane, with its corresponding matched filter function to obtain a matched signal, as shown in equation (13):

[0138]

[0139] Where I(y,h,p) represents all pixels in the p-th azimuth tangent plane, i.e., the p-th range-height two-dimensional plane. This represents the result after matched filtering.

[0140] S524. Perform distance compression processing on the matching signal, that is, perform inverse Fourier transform on the matching signal in the distance tangent. Specifically, as shown in equation (14):

[0141]

[0142] S525. Perform coherent superposition processing on the peak values ​​of the distance compression signal of each pixel after transformation to obtain the superposition result of the peak values ​​of the distance compression signals of all pixels in the azimuth section.

[0143] Specifically, each pixel in the p-th distance-height two-dimensional plane after distance compression is taken, and then the peak value of the signal of each pixel is coherently superimposed to obtain the superposition result of the peak values ​​of the signals of all pixels in the distance-height two-dimensional plane, denoted as I(y,h,p). The superposition result I(q,p) of the peak value of one pixel signal is shown in Equation (15):

[0144]

[0145] Where I'(q,p) represents the peak signal value of the q-th pixel in the p-th distance-height 2D plane, and N h This indicates the number of sampling points along the height.

[0146] S53. Process the second count variable and compare the second count variable with the second decomposition quantity, wherein the second count variable corresponds to the position of the two-dimensional image in the azimuth section, and the second decomposition quantity is the number of azimuth grid cells;

[0147] S54. Based on the second comparison result, determine whether to continue the step of constructing the matched filter matrix until the three-dimensional image of the target point is obtained.

[0148] As an optional embodiment, such as Figure 9 As shown, the step of determining whether to continue constructing the matched filter matrix based on the second comparison result until the three-dimensional image of the target point is obtained includes:

[0149] S541. If the processed second count variable falls within the range of the second decomposition quantity, then continue with the step of constructing the matched filter matrix. Specifically, make the second count variable p+1, if p≤N x Continue with step S42.

[0150] S542. If the processed second count variable does not fall within the range of the second decomposition quantity, then stop executing the step of constructing the matched filter matrix to obtain the three-dimensional image corresponding to the target point.

[0151] Specifically, when the second count variable P is greater than the second decomposition quantity, step S42 is stopped, where N x This represents the number of azimuth grid cells, which is the number of all two-dimensional planes along the azimuth tangent in the range and height directions. After the process is completed, the three-dimensional image corresponding to the three-dimensional echo data of the sampling point is obtained.

[0152] One of the objectives of this application is to provide a three-dimensional imaging system for millimeter-wave radar sparse data, such as... Figure 10 As shown, it includes:

[0153] The first acquisition module 100 is configured to acquire three-dimensional echo data of sampling points, wherein the sampling points are used to scan target points;

[0154] Decomposition module 200 is configured to decompose the three-dimensional echo data in the height-oriented tangent to obtain two-dimensional echo data of layer Nh.

[0155] The construction module 300 is configured to construct observation equations based on the two-dimensional echo data;

[0156] The second acquisition module 400 is configured to acquire Nh two-dimensional images based on the observation equation and using the first algorithm.

[0157] The third acquisition module 500 is configured to perform coherent superposition processing of signal peaks on each pixel in the azimuth section of the Nh images of the two-dimensional image to obtain a three-dimensional image of the target point.

[0158] The millimeter-wave radar sparse data three-dimensional imaging system described in this application embodiment can implement the steps of the millimeter-wave radar sparse data three-dimensional imaging method mentioned in any embodiment of this application through its configured functional modules. Therefore, the implementation of the millimeter-wave radar sparse data three-dimensional imaging system provided in this application embodiment can refer to the implementation of the millimeter-wave radar sparse data three-dimensional imaging method provided in this application, and will not be repeated here.

[0159] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for three-dimensional imaging of sparse data from millimeter-wave radar, characterized in that, include: Acquire three-dimensional echo data of sampling points, wherein the sampling points are used to scan the target point; The three-dimensional echo data is decomposed in the height-oriented section to obtain... Layer 2D echo data; An observation equation is constructed based on the two-dimensional echo data; Based on the observation equation, the first algorithm is used to obtain... Two-dimensional image; right The two-dimensional image is subjected to coherent superposition of signal peaks for each pixel along the azimuth section to obtain a three-dimensional image of the target point, which includes: For each pixel in the azimuth section of each of the two-dimensional images, coherent superposition of signal peaks is performed; Construct a corresponding matched filter matrix for each of the processed azimuth tangents; Process the second count variable and compare the second count variable with the second decomposition quantity, wherein the second count variable corresponds to the position of the two-dimensional image in the azimuth section, and the second decomposition quantity is the number of azimuth grid cells; Based on the second comparison result, determine whether to continue executing the step of constructing the matched filter matrix until a three-dimensional image of the target point is obtained; The step of constructing a corresponding matched filter matrix for each processed azimuth tangent includes: Construct the distance matrix from the sampling point to each of the azimuth tangents; Process the distance matrix to obtain the matched filter matrix; Multiply each pixel in the azimuth section with its corresponding matched filter matrix to obtain the matched signal; The matching signal is subjected to distance compression processing; The peak values ​​of the distance compression signal of each pixel after transformation are coherently superimposed to obtain the superposition result of the peak values ​​of the distance compression signal of all pixels in the azimuth section. The step of determining whether to continue constructing the matched filter matrix based on the second comparison result until the three-dimensional image of the target point is obtained includes: If the processed second count variable falls within the range of the second decomposition quantity, then continue with the step of constructing the matched filter matrix; If the processed second count variable does not fall within the range of the second decomposition quantity, then the step of constructing the matched filter matrix is ​​stopped to obtain the three-dimensional image corresponding to the target point.

2. The method for three-dimensional imaging of sparse data from millimeter-wave radar according to claim 1, characterized in that, The acquisition of the three-dimensional echo data of the sampling points includes: Define a planar scanning 3D imaging model, where the coordinates of the target point are... The coordinates of the sampling point are ; Based on the planar scanning 3D imaging model, the distance between the sampling point and the geometric center of the target point is calculated.

3. The method for three-dimensional imaging of sparse data from millimeter-wave radar according to claim 1, characterized in that, The construction of the observation equation based on the two-dimensional echo data includes: Based on a preset step size, the preset region is divided into imaging region grids, and the division results are obtained; Based on the partitioning results, a distance matrix is ​​constructed; Based on the distance matrix, construct the observation matrix; Based on the division results, an imaging region grid matrix is ​​constructed; Based on the two-dimensional echo data, a two-dimensional echo matrix is ​​constructed; The observation equation is constructed based on the distance matrix, the observation matrix, the imaging region raster matrix, and the two-dimensional echo matrix.

4. The millimeter-wave radar sparse data three-dimensional imaging method according to claim 3, characterized in that, The process of dividing a preset region into imaging region grids based on a preset step size and obtaining the division results includes: The preset region in the distance direction is divided into imaging grids by a preset first step length to obtain multiple distance units; The preset region in the azimuth direction is divided into an imaging grid by a preset second step length to obtain multiple azimuth units; The segmentation result is obtained based on multiple distance units and multiple orientation units.

5. The millimeter-wave radar sparse data three-dimensional imaging method according to claim 3, characterized in that, Based on the observation equation, the first algorithm is used to obtain... Two-dimensional images, including: The scattering information column vector of the imaging region grid matrix is ​​calculated using the first formula, and the scattering information column vector is rearranged into a scattering information matrix to obtain a two-dimensional image. The first count variable is processed, and the processed first count variable is compared with the first decomposition number, wherein the first count variable corresponds to the position of the two-dimensional image in the height section, and the first decomposition number is the decomposition number of the three-dimensional echo data; Based on the first comparison result, determine whether to continue executing the step of calculating the scattering information column vector until the result is obtained. A two-dimensional image.

6. The method for three-dimensional imaging of sparse data from millimeter-wave radar according to claim 5, characterized in that, The step of determining whether to continue calculating the scattering information column vector based on the first comparison result, until the result is obtained... Two-dimensional images, including: If the processed first count variable falls within the range of the first decomposition quantity, then continue to execute the step of calculating the scattering information column vector; If the processed first count variable does not fall within the range of the first decomposition quantity, then stop executing the step of calculating the scattering information column vector to obtain the corresponding first decomposition quantity. A two-dimensional image.

7. A millimeter-wave radar sparse data three-dimensional imaging system, characterized in that, include: The first acquisition module is configured to acquire three-dimensional echo data of sampling points, wherein the sampling points are used to scan the target point; The decomposition module is configured to decompose the three-dimensional echo data in the height-oriented section to obtain... Layer 2D echo data; The construction module is configured to construct observation equations based on the two-dimensional echo data; The second acquisition module is configured to acquire data based on the observation equation using the first algorithm. Two-dimensional image; The third acquisition module is configured to... The two-dimensional image is subjected to coherent superposition of signal peaks for each pixel along the azimuth section to obtain a three-dimensional image of the target point, which includes: For each pixel in the azimuth section of each of the two-dimensional images, coherent superposition of signal peaks is performed; Construct a corresponding matched filter matrix for each of the processed azimuth tangents; Process the second count variable and compare the second count variable with the second decomposition quantity, wherein the second count variable corresponds to the position of the two-dimensional image in the azimuth section, and the second decomposition quantity is the number of azimuth grid cells; Based on the second comparison result, determine whether to continue executing the step of constructing the matched filter matrix until a three-dimensional image of the target point is obtained; The step of constructing a corresponding matched filter matrix for each processed azimuth tangent includes: Construct the distance matrix from the sampling point to each of the azimuth tangents; Process the distance matrix to obtain the matched filter matrix; Multiply each pixel in the azimuth section with its corresponding matched filter matrix to obtain the matched signal; The matching signal is subjected to distance compression processing; The peak values ​​of the distance compression signal of each pixel after transformation are coherently superimposed to obtain the superposition result of the peak values ​​of the distance compression signal of all pixels in the azimuth section. The step of determining whether to continue constructing the matched filter matrix based on the second comparison result until the three-dimensional image of the target point is obtained includes: If the processed second count variable falls within the range of the second decomposition quantity, then continue with the step of constructing the matched filter matrix; If the processed second count variable does not fall within the range of the second decomposition quantity, then the step of constructing the matched filter matrix is ​​stopped to obtain the three-dimensional image corresponding to the target point.

Images