A point-trace aggregation method for extended targets in millimeter-wave radar based on morphology

Through morphology-based methods, millimeter-wave radar point trace condensation is solved, and the problems of low efficiency and poor accuracy in the existing technology are achieved, efficient and high-precision point trace condensation is suitable for radar target classification and data correlation.

CN118884356BActive Publication Date: 2025-01-03ANHUI FALCON WAVE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411324518.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-23
Publication Date
2025-01-03
Estimated Expiration
2044-09-23

AI Technical Summary

Technical Problem

In the prior art, the point trace aggregation efficiency in millimeter-wave radar data processing is low, and it is difficult to give an accurate profile of the detection target.

Method used

Using a morphology-based method, radar point trace data is obtained through the CFAR algorithm, polar coordinate system storage, expansion operation, point trace condensation and corrosion operations are performed, and the agglomeration center point and condensation area are determined.

Benefits of technology

Low algorithm complexity and efficient point trace aggregation are achieved, and the output aggregation area is more in line with the real shape of the detection target, which helps the radar target classification and data correlation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118884356B_ABST
    Figure CN118884356B_ABST
Patent Text Reader

Abstract

The invention relates to millimeter wave radar data processing, and in particular to a point trace condensation method for a millimeter wave radar extended target based on morphology. The method comprises the following steps: obtaining radar point trace data of a millimeter wave radar based on a CFAR algorithm, and storing the radar point trace data in a radar point trace visual graph based on a polar coordinate system; performing a preset number of expansion operations on each radar point trace in the radar point trace visual graph, and judging whether there is an overlapping area between the expansion ranges of each radar point trace; judging whether two radar point traces forming the overlapping area can be condensed for the overlapping area, and if the two radar point traces can be condensed, setting the overlapping area as a connection point, and otherwise setting the overlapping area as an isolation point; performing an erosion operation on the radar point trace and its expansion point according to the connection point; performing point trace condensation, and determining a condensation center point and a condensation area; and the technical solution provided by the invention can overcome the defects of low efficiency and difficulty in providing an accurate detection target outline.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to millimeter wave radar data processing, and in particular to a point trace condensation method of a millimeter wave radar extended target based on morphology. Background Art

[0002] With the rapid development of millimeter-wave radar technology, the range resolution and angle resolution of radar are gradually improving. A single motor vehicle detected by radar will return multiple target points, thus occupying multiple range cells and angle cells. These detection targets occupying multiple cells are called extended targets.

[0003] Point trace aggregation is an important part of radar data processing technology and one of the key contents of radar data processing. Through point trace aggregation, it is determined which returned target points belong to the same detection target, and the center of mass and shape contour of the detection target are estimated, which is the basis for subsequent target tracking and target classification.

[0004] In the prior art, point agglomeration mostly uses the density-based DBSCAN algorithm, which has the advantages of being applicable to any shape and being highly robust to noise and outliers. However, the DBSCAN algorithm uses a recursive exhaustive method to search the neighborhood, which makes the overall time complexity high and it is difficult to give an accurate outline of the detected target. Summary of the invention

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides a point trace aggregation method for extended targets of millimeter-wave radar based on morphology, which can effectively overcome the defects of the prior art such as low efficiency and difficulty in giving accurate detection target contour.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0007] A point trace condensation method for extended targets of millimeter wave radar based on morphology comprises the following steps:

[0008] S1. Obtaining radar point trace data of a millimeter wave radar based on a CFAR algorithm, and storing the radar point trace data in a radar point trace visual graph based on a polar coordinate system;

[0009] S2, performing a preset number of expansion operations on each radar point trace in the radar point trace visual graph to determine whether there is an overlapping area between the expansion ranges of each radar point trace;

[0010] S3, for the overlapping area, determine whether the two radar traces forming the overlapping area can be condensed, if the two radar traces can be condensed, set the overlapping area as a connection point, otherwise set the overlapping area as an isolation point;

[0011] S4, performing corrosion operation on the radar trace and its expansion point according to the connection point;

[0012] S5. Perform point condensation and determine the condensation center point and condensation area.

[0013] Preferably, in S1, radar trace data of the millimeter wave radar is obtained based on the CFAR algorithm, and the radar trace data is stored in a radar trace visual graph based on a polar coordinate system, including:

[0014] S11. Obtaining radar trace data of millimeter-wave radar based on CA-CFAR algorithm P ={ p 1, p 2,…, p N}, among which i The radar trace is , p ir Radar trace p i The number of distance dimensions, Radar trace p i The number of angle dimensions, p iv Radar trace p i The number of velocity dimensions, p iRCS Radar trace p i of RCS value, , N The radar trace data after constant false alarm detection P The number of radar traces in ;

[0015] S12, according to the distance dimension number and angle dimension number of each radar point trace, the velocity dimension number and RCS The value is stored in the corresponding position in the radar trace visual graph;

[0016] Among them, the horizontal axis of the radar trace visualization graph represents the number of angle dimensions, and the corresponding maximum number of angle dimensions of radar detection is AngleN , the vertical coordinate of the radar trace visualization graph represents the number of distance dimensions, and the corresponding maximum distance dimension of radar detection is RangeN .

[0017] Preferably, in S2, before performing a preset number of expansion operations on each radar trace in the radar trace visual graph and determining whether there is an overlapping area between the expansion ranges of each radar trace, the following steps are included:

[0018] S21, determining the number of expansion operations according to the detection target, distance resolution, and angle resolution of the millimeter wave radar;

[0019] Among them, when the detection target is a motor vehicle, and the distance resolution of the millimeter wave radar is 1.34m, and the angle resolution is 0.2°, it is determined to perform two expansion operations on each radar point trace in the radar point trace visible graph.

[0020] Preferably, in S2, performing a preset number of expansion operations on each radar point trace in the radar point trace visual graph to determine whether there is an overlapping area between the expansion ranges of each radar point trace includes:

[0021] S22, performing an expansion operation on each radar point trace in the radar point trace visual graph:

[0022] Based on the position of the radar trace, the expansion operation is performed in eight directions, namely, upper left, upper, upper right, right, lower right, lower, lower left, and left, in a clockwise direction to obtain the first expansion point of the radar trace. When the following two situations occur, the expansion operation in the corresponding direction is stopped:

[0023] 1) The first expansion point obtained is outside the radar point trace visible graph;

[0024] 2) There are already radar traces in a certain direction;

[0025] Among them, the number of velocity dimensions of the first expansion point of the radar trace and RCS The value is the same as that of the radar trace;

[0026] S23, performing a secondary expansion operation on each radar point trace in the radar point trace visual graph:

[0027] Based on the position of the first expansion point of the radar trace, the expansion operation is performed in eight directions, namely, upper left, upper, upper right, right, lower right, lower, lower left, and left, in a clockwise direction to obtain the second expansion point of the radar trace. When the following three situations occur, the expansion operation in the corresponding direction is stopped:

[0028] 1) The obtained secondary expansion point is outside the radar point trace visible graph;

[0029] 2) There is already an expansion point in a certain direction;

[0030] 3) There are already radar traces in a certain direction;

[0031] S24, judging whether there is an overlapping area between the expansion ranges of the radar traces, if there is no overlapping area, no operation is performed, otherwise the process goes to S3;

[0032] The expansion range of the radar trace includes the radar trace itself and the coverage of its first expansion point and second expansion point.

[0033] Preferably, in S3, for the overlapping area, it is determined whether the two radar traces forming the overlapping area can be condensed, if the two radar traces can be condensed, the overlapping area is set as a connection point, otherwise the overlapping area is set as an isolation point, including:

[0034] S31, determining radar traces forming overlapping areas p i 、Radar traces p j The conditions for determining whether point condensation can be performed are:

[0035] ;

[0036] S32: If radar traces in overlapping areas are formed p i 、Radar traces p j If the determination condition in S31 is met, the radar traces forming the overlapping area are determined. p i 、Radar traces p j It is possible to perform point condensation and set the overlapping area as a connection point, otherwise it is set as an isolation point;

[0037] S33, calculate the number and speed dimensions of each connection point in the overlapping area RCS value:

[0038] For radar traces p i 、Radar traces p j Forming connection points in overlapping areas p k , the connection point is calculated using the following formula p k The number of speed dimensions p kv :

[0039] ;

[0040] For radar traces p i 、Radar traces p j Forming connection points in overlapping areas p k , the connection point is calculated using the following formula p k RCS value p kRCS :

[0041] ;

[0042] in, p jr Radar trace p j The number of distance dimensions, Radar trace p j The number of angle dimensions, p jv Radar trace p j The number of velocity dimensions, p jRCS Radar trace p j of RCS value, p kr For connection point p k The number of distance dimensions, For connection point p k The number of angle dimensions, is the floor function.

[0043] Preferably, in S4, the radar trace and its expansion point are corroded according to the connection point, including:

[0044] 1) For radar traces that do not have an overlapping area with the expansion range of other radar traces, the radar trace is marked as a stray trace, and the stray trace and its first and second expansion points are deleted from the radar trace visual graph;

[0045] 2) When there is an overlapping area between the expansion ranges of the radar traces, and the overlapping area is a connection point, some of the first expansion points and all the second expansion points of the radar traces are deleted from the radar trace visual graph according to the first corrosion operation rule;

[0046] Among them, the first corrosion operation rule includes:

[0047] If the point is a secondary expansion point, the point information is deleted from the radar point visible graph;

[0048] If the point is a primary expansion point and there are no connecting points in the adjacent positions of the point in the four directions of up, down, left and right, the point information is deleted from the radar point visual graph;

[0049] If the point trace is a radar point trace, and there are no connected points in the adjacent positions of the eight directions of the point trace, namely, upper left, upper right, right, lower right, lower, lower left, and left, then delete the point trace information from the radar point trace visual graph;

[0050] 3) When there is an overlapping area between the expansion ranges of the radar traces, and the overlapping area is an isolated point, the radar traces and some of their first expansion points and all their second expansion points are deleted from the radar trace visual graph according to the second corrosion operation rule;

[0051] The second corrosion operation rule includes:

[0052] If the point is a secondary expansion point, the point information is deleted from the radar point visible graph;

[0053] If the point is a primary expansion point and there are no connecting points in the adjacent positions of the point in the four directions of up, down, left and right, the point information is deleted from the radar point visual graph;

[0054] If the point trace is a radar point trace, and there are no connected points in the adjacent positions of the eight directions of the point trace, namely, upper left, upper right, right, lower right, lower, lower left, and left, then delete the point trace information from the radar point trace visual graph;

[0055] If the point is an isolated point, the point information is retained and used for segmentation judgment between targets.

[0056] Preferably, performing point condensation in S5 and determining the condensation center point and condensation area include:

[0057] S51, use the following formula to calculate the center point of condensation p C The number of distance dimensions p Cr , number of angle dimensions , the number of speed dimensions p Cv and RCS value p CRCS :

[0058] ;

[0059] ;

[0060] ;

[0061] ;

[0062] in, p l are the points participating in point condensation, that is, the points retained in the radar point visible graph after the corrosion operation. p lr Points participating in the aggregation of points p l The number of distance dimensions, Points participating in the aggregation of pointsp l The number of angle dimensions, p lv Points participating in the aggregation of points p l The number of velocity dimensions, p lRCS Points participating in the aggregation of points p l of RCS value, , L is the number of points participating in the point condensation;

[0063] S52, determining the edge of the condensed area according to the condensed area edge judgment rule, thereby obtaining the condensed area;

[0064] The condensed area edge judgment rules include:

[0065] If the side outside the corresponding position of the point trace participating in the point trace condensation in the radar point trace visible graph is an isolation point or a blank area, then the side is the edge of the condensation area.

[0066] Compared with the prior art, the point trace condensation method of millimeter wave radar extended target based on morphology provided by the present invention has the following beneficial effects:

[0067] 1) The morphological method in image processing was introduced to complete the radar point condensation with low algorithm complexity, which has the advantages of good real-time performance and high point condensation accuracy;

[0068] 2) Based on operations such as dilation and erosion, the output condensed area is made to fit the real shape of the detected target more closely, which is helpful for radar target classification and data association in the process of radar video data fusion. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0070] Figure 1 It is a schematic diagram of the process of the present invention;

[0071] Figure 2 A schematic diagram of storing radar traces in a radar trace visual graph based on a polar coordinate system;

[0072] Figure 3A schematic diagram of performing a dilation operation on two radar traces in a radar trace visual graph;

[0073] Figure 4 For Figure 3 Schematic diagram of the secondary expansion operation of the two radar traces in FIG.

[0074] Figure 5 It is a schematic diagram showing that there is no overlapping area between the expansion ranges of two radar traces in the radar trace visual graph;

[0075] Figure 6 A schematic diagram showing an overlapping area between expansion ranges of two radar traces in the radar trace visual graph;

[0076] Figure 7 for Figure 6 The overlapping area in is a schematic diagram of the connection points;

[0077] Figure 8 for Figure 6 The overlapping area in the figure is a schematic diagram of the isolation points;

[0078] Fig. 9 It is a schematic diagram showing the existence of stray traces in the radar trace visual graph;

[0079] Fig.10 for Fig. 9 Schematic diagram after the etching operation;

[0080] Fig.11 For Figure 7 Schematic diagram after the etching operation;

[0081] Fig.12 For Figure 8 Schematic diagram after the etching operation;

[0082] Fig.13 It is a schematic diagram of multiple radar traces in the radar trace visual graph after the expansion operation and whether the trace condensation can be performed;

[0083] Fig.14 For Fig.13 Schematic diagram after the etching operation;

[0084] Fig.15 for Fig.14 Schematic diagram of the condensation center point and condensation area. DETAILED DESCRIPTION

[0085] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0086] A point condensation method for extended targets of millimeter wave radar based on morphology, such as Figure 1 As shown, S1, based on the CFAR algorithm, the radar trace data of the millimeter wave radar is obtained, and the radar trace data is stored in the radar trace visual graph based on the polar coordinate system, specifically including:

[0087] S11. Obtaining radar trace data of millimeter-wave radar based on CA-CFAR algorithm P ={ p 1, p 2,…, p N}, among which i The radar trace is , p ir Radar trace p i The number of distance dimensions, Radar trace p i The number of angle dimensions, p iv Radar trace p i The number of velocity dimensions, p iRCS Radar trace p i of RCS value, , N The radar trace data after constant false alarm detection P The number of radar traces in ;

[0088] S12, according to the distance dimension number and angle dimension number of each radar point trace, the velocity dimension number and RCS The value is stored in the corresponding position in the radar trace visual graph;

[0089] Among them, the horizontal axis of the radar trace visualization graph represents the number of angle dimensions, and the corresponding maximum number of angle dimensions of radar detection is AngleN , the vertical coordinate of the radar trace visualization graph represents the number of distance dimensions, and the corresponding maximum distance dimension of radar detection is RangeN .

[0090] like Figure 2 As shown, for radar point data P ={ p 1, p 2,…, p N} i Radar traces ,like p ir =6, =5, then the number of speed dimensions p iv and RCS value p iRCS Stored to the corresponding position in the radar trace visual graph, that is, Figure 2 middle Location.

[0091] S2. Perform a preset number of expansion operations on each radar point trace in the radar point trace visual graph to determine whether there is an overlapping area between the expansion ranges of each radar point trace.

[0092] ① Performing a preset number of expansion operations on each radar point trace in the radar point trace visual graph and determining whether there is an overlapping area between the expansion ranges of each radar point trace, including:

[0093] S21, determining the number of expansion operations according to the detection target, distance resolution, and angle resolution of the millimeter wave radar;

[0094] Among them, when the detection target is a motor vehicle, and the distance resolution of the millimeter wave radar is 1.34m, and the angle resolution is 0.2°, it is determined to perform two expansion operations on each radar point trace in the radar point trace visible graph.

[0095] ② Perform a preset number of expansion operations on each radar trace in the radar trace visual graph to determine whether there is an overlapping area between the expansion ranges of each radar trace, including:

[0096] S22, performing an expansion operation on each radar point trace in the radar point trace visual graph:

[0097] Based on the position of the radar trace, the expansion operation is performed in 8 directions, namely, upper left, upper, upper right, right, lower right, lower, lower left, and left, in a clockwise direction to obtain the first expansion point of the radar trace (such as Figure 3 As shown, is the radar trace, is an expansion point of the radar trace). When the following two situations occur, the expansion operation in the corresponding direction is stopped:

[0098] 1) The first expansion point obtained is outside the radar point trace visible graph;

[0099] 2) There are already radar traces in a certain direction;

[0100] Among them, the number of velocity dimensions of the first expansion point of the radar trace and RCS The value is the same as that of the radar trace;

[0101] S23, performing a secondary expansion operation on each radar point trace in the radar point trace visual graph:

[0102] Based on the position of the first expansion point of the radar trace, the expansion operation is performed in eight directions, namely, upper left, upper, upper right, right, lower right, lower, lower left, and left, in a clockwise direction to obtain the second expansion point of the radar trace (such as Figure 4 As shown, is the radar trace, is a first expansion point of the radar trace, is the secondary expansion point of the radar trace). When the following three situations occur, the expansion operation in the corresponding direction is stopped:

[0103] 1) The obtained secondary expansion point is outside the radar point trace visible graph;

[0104] 2) There is already an expansion point in a certain direction;

[0105] 3) There are already radar traces in a certain direction;

[0106] S24, determine whether there is an overlapping area between the expansion ranges of the radar traces (such as Figure 5 and Figure 6 As shown), if there is no overlapping area, no operation is performed, otherwise, the process goes to S3;

[0107] The expansion range of the radar trace includes the radar trace itself and the coverage of its first expansion point and second expansion point.

[0108] S3. For the overlapping area, determine whether the two radar traces forming the overlapping area can be condensed. If the two radar traces can be condensed, set the overlapping area as a connection point; otherwise, set the overlapping area as an isolation point, specifically including:

[0109] S31, determining radar traces forming overlapping areas p i 、Radar traces p j The conditions for determining whether point condensation can be performed are:

[0110] ;

[0111] S32: If radar traces in overlapping areas are formed pi 、Radar traces p j If the determination condition in S31 is met, the radar traces forming the overlapping area are determined. p i 、Radar traces p j It is possible to condense points and set the overlapping area as the connection point (such as Figure 7 As shown, is the radar trace, is a first expansion point of the radar trace, is the secondary expansion point of the radar trace, is a connection point), otherwise the overlapping area is set as an isolation point (such as Figure 8 As shown, is the radar trace, is a primary expansion point of the radar trace, is the secondary expansion point of the radar trace, for isolation points);

[0112] S33, calculate the number and speed dimensions of each connection point in the overlapping area RCS value:

[0113] For radar traces p i 、Radar traces p j Forming connection points in overlapping areas p k , the connection point is calculated using the following formula p k The number of speed dimensions p kv :

[0114] ;

[0115] For radar traces p i 、Radar traces p j Forming connection points in overlapping areas p k , the connection point is calculated using the following formula p k RCS value p kRCS :

[0116] ;

[0117] in, p jr Radar trace p j The number of distance dimensions, Radar trace p j The number of angle dimensions, p jv Radar trace p j The number of velocity dimensions, p jRCS Radar trace p j of RCS value, p kr For connection point p k The number of distance dimensions, For connection point p k The number of angle dimensions, is the floor function.

[0118] S4. Corrosion operation is performed on the radar trace and its expansion point according to the connection point, specifically including:

[0119] 1) For radar traces that do not overlap with the expansion range of other radar traces, mark the radar trace as a stray trace (such as Fig. 9 ), and delete the stray trace and its first expansion point and second expansion point (as shown in Fig.10 shown);

[0120] 2) If there is an overlapping area between the expansion ranges of the radar traces, and the overlapping area is a connection point, some of the first expansion points and all the second expansion points of the radar traces are deleted from the radar trace visual graph according to the first corrosion operation rule (such as Fig.11 shown);

[0121] Among them, the first corrosion operation rule includes:

[0122] If the point is a secondary expansion point, the point information is deleted from the radar point visible graph;

[0123] If the point is a primary expansion point and there are no connecting points in the adjacent positions of the point in the four directions of up, down, left and right, the point information is deleted from the radar point visual graph;

[0124] If the point trace is a radar point trace, and there are no connected points in the adjacent positions of the eight directions of the point trace, namely, upper left, upper right, right, lower right, lower, lower left, and left, then delete the point trace information from the radar point trace visual graph;

[0125] 3) If there is an overlapping area between the expansion ranges of the radar traces, and the overlapping area is an isolated point, the radar traces and some of their first expansion points and all their second expansion points are deleted from the radar trace visual graph according to the second corrosion operation rule (such as Fig.12 shown);

[0126] The second corrosion operation rule includes:

[0127] If the point is a secondary expansion point, the point information is deleted from the radar point visible graph;

[0128] If the point is a primary expansion point and there are no connecting points in the adjacent positions of the point in the four directions of up, down, left and right, the point information is deleted from the radar point visual graph;

[0129] If the point trace is a radar point trace, and there are no connected points in the adjacent positions of the eight directions of the point trace, namely, upper left, upper right, right, lower right, lower, lower left, and left, then delete the point trace information from the radar point trace visual graph;

[0130] If the point is an isolated point, the point information is retained and used for segmentation judgment between targets.

[0131] In the technical solution of this application, Fig.11 and Fig.12 All of them are the results obtained after the expansion and corrosion operations are performed when there are only two radar traces in the radar trace visual graph. When there are multiple radar traces in the radar trace visual graph, the results obtained after the expansion operation and the judgment of whether the trace can be condensed are performed according to the above S2 and S3 are as follows Fig.13 As shown, the corrosion operation is performed according to the above S4 and the result is as follows Fig.14 shown.

[0132] S5, performing point condensation and determining the condensation center point and condensation area, specifically including:

[0133] S51, use the following formula to calculate the center point of condensation p C (like Fig.15 As shown, Fig.15 for Fig.14 Schematic diagram of the condensation center point and condensation area in the figure, where The number of distance dimensions of the center point p Cr , number of angle dimensions , the number of speed dimensions p Cv and RCS value p CRCS :

[0134] ;

[0135] ;

[0136] ;

[0137] ;

[0138] in, p l are the points participating in point condensation, that is, the points retained in the radar point visible graph after the corrosion operation. p lr Points participating in the aggregation of points p l The number of distance dimensions, Points participating in the aggregation of points p l The number of angle dimensions, p lv Points participating in the aggregation of points p l The number of velocity dimensions, p lRCS Points participating in the aggregation of points p l of RCS value, , L is the number of points participating in the point condensation;

[0139] S52, determining the edge of the condensed area according to the condensed area edge judgment rule, thereby obtaining the condensed area (such as Fig.15 As shown, Fig.15 for Fig.14 Schematic diagram of the condensation center point and condensation area, where the area surrounded by the dotted line is the condensation area);

[0140] The condensed area edge judgment rules include:

[0141] If the side outside the corresponding position of the point trace participating in the point trace condensation in the radar point trace visible graph is an isolation point or a blank area, then the side is the edge of the condensation area.

[0142] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A point trace condensation method for extended targets of millimeter wave radar based on morphology, characterized by: The following steps are involved: S1. Obtaining radar point trace data of a millimeter wave radar based on a CFAR algorithm, and storing the radar point trace data in a radar point trace visual graph based on a polar coordinate system; S2, performing a preset number of expansion operations on each radar point trace in the radar point trace visual graph to determine whether there is an overlapping area between the expansion ranges of each radar point trace; S3, for the overlapping area, determine whether the two radar traces forming the overlapping area can be condensed, if the two radar traces can be condensed, set the overlapping area as a connection point, otherwise set the overlapping area as an isolation point; S4, performing corrosion operation on the radar trace and its expansion point according to the connection point; S5, performing point condensation, and determining the condensation center point and condensation area; In S4, the radar traces and their expansion points are corroded according to the connection points, including: 1) For radar traces that do not have an overlapping area with the expansion range of other radar traces, the radar trace is marked as a stray trace, and the stray trace and its first and second expansion points are deleted from the radar trace visual graph; 2) When there is an overlapping area between the expansion ranges of the radar traces, and the overlapping area is a connection point, some of the first expansion points and all the second expansion points of the radar traces are deleted from the radar trace visual graph according to the first corrosion operation rule; Among them, the first corrosion operation rule includes: If the point is a secondary expansion point, the point information is deleted from the radar point visible graph; If the point is a primary expansion point and there are no connecting points in the adjacent positions of the point in the four directions of up, down, left and right, the point information is deleted from the radar point visual graph; If the point trace is a radar point trace, and there are no connected points in the adjacent positions of the eight directions of the point trace, namely, upper left, upper right, right, lower right, lower, lower left, and left, then delete the point trace information from the radar point trace visual graph; 3) When there is an overlapping area between the expansion ranges of the radar traces, and the overlapping area is an isolated point, the radar traces and some of their first expansion points and all their second expansion points are deleted from the radar trace visual graph according to the second corrosion operation rule; The second corrosion operation rule includes: If the point is a secondary expansion point, the point information is deleted from the radar point visible graph; If the point is a primary expansion point and there are no connecting points in the adjacent positions of the point in the four directions of up, down, left and right, the point information is deleted from the radar point visual graph; If the point trace is a radar point trace, and there are no connected points in the adjacent positions of the eight directions of the point trace, namely, upper left, upper right, right, lower right, lower, lower left, and left, then delete the point trace information from the radar point trace visual graph; If the point is an isolated point, the point information is retained and used for segmentation judgment between targets.

2. The point trace aggregation method for extended targets of millimeter wave radar based on morphology according to claim 1 is characterized in that: In S1, the radar trace data of the millimeter-wave radar is obtained based on the CFAR algorithm, and the radar trace data is stored in the radar trace visual graph based on the polar coordinate system, including: S11. Obtaining radar trace data of millimeter-wave radar based on CA-CFAR algorithm P ={ p 1, p 2,…, p N }, among which i The radar trace is , p ir Radar trace p i The number of distance dimensions, Radar trace p i The number of angle dimensions, p iv Radar trace p i The number of velocity dimensions, p iRCS Radar trace p i of RCS value, , N The radar trace data after constant false alarm detection P The number of radar traces in ; S12, according to the distance dimension number and angle dimension number of each radar point trace, the velocity dimension number and RCS The value is stored in the corresponding position in the radar trace visual graph; Among them, the horizontal axis of the radar trace visualization graph represents the number of angle dimensions, and the corresponding maximum number of angle dimensions of radar detection is AngleN , the vertical coordinate of the radar trace visualization graph represents the number of distance dimensions, and the corresponding maximum distance dimension of radar detection is RangeN .

3. The point trace aggregation method for extended targets of millimeter wave radar based on morphology according to claim 2 is characterized in that: In S2, before performing a preset number of expansion operations on each radar point trace in the radar point trace visual graph and determining whether there is an overlapping area between the expansion ranges of each radar point trace, the method includes: S21, determining the number of expansion operations according to the detection target, distance resolution, and angle resolution of the millimeter wave radar; Among them, when the detection target is a motor vehicle, and the distance resolution of the millimeter wave radar is 1.34m, and the angle resolution is 0.2°, it is determined to perform two expansion operations on each radar point trace in the radar point trace visible graph.

4. The point trace aggregation method for extended targets of millimeter wave radar based on morphology according to claim 3 is characterized in that: In S2, a preset number of expansion operations are performed on each radar point trace in the radar point trace visual graph to determine whether there is an overlapping area between the expansion ranges of each radar point trace, including: S22, performing an expansion operation on each radar point trace in the radar point trace visual graph: Based on the position of the radar trace, the expansion operation is performed in eight directions, namely, upper left, upper, upper right, right, lower right, lower, lower left, and left, in a clockwise direction to obtain the first expansion point of the radar trace. When the following two situations occur, the expansion operation in the corresponding direction is stopped: 1) The first expansion point obtained is outside the radar point trace visible graph; 2) There are already radar traces in a certain direction; Among them, the number of velocity dimensions of the first expansion point of the radar trace and RCS The value is the same as that of the radar trace; S23, performing a secondary expansion operation on each radar point trace in the radar point trace visual graph: Based on the position of the first expansion point of the radar trace, the expansion operation is performed in eight directions, namely, upper left, upper, upper right, right, lower right, lower, lower left, and left, in a clockwise direction to obtain the second expansion point of the radar trace. When the following three situations occur, the expansion operation in the corresponding direction is stopped: 1) The obtained secondary expansion point is outside the radar point trace visible graph; 2) There is already an expansion point in a certain direction; 3) There are already radar traces in a certain direction; S24, judging whether there is an overlapping area between the expansion ranges of the radar traces, if there is no overlapping area, no operation is performed, otherwise the process goes to S3; The expansion range of the radar trace includes the radar trace itself and the coverage of its first expansion point and second expansion point.

5. The point trace aggregation method for extended targets of millimeter wave radar based on morphology according to claim 4 is characterized in that: In S3, for the overlapping area, it is determined whether the two radar traces forming the overlapping area can be condensed. If the two radar traces can be condensed, the overlapping area is set as a connection point, otherwise the overlapping area is set as an isolation point, including: S31, determining radar traces forming overlapping areas p i 、Radar traces p j The conditions for determining whether point condensation can be performed are: ; S32: If radar traces in overlapping areas are formed p i 、Radar traces p j If the determination condition in S31 is met, the radar traces forming the overlapping area are determined. p i 、Radar traces p j It is possible to perform point condensation and set the overlapping area as a connection point, otherwise it is set as an isolation point; S33, calculate the number and speed dimensions of each connection point in the overlapping area RCS value: For radar traces p i 、Radar traces p j Forming connection points in overlapping areas p k , the connection point is calculated using the following formula p k The number of speed dimensions p kv : ; For radar traces p i 、Radar traces p j Forming connection points in overlapping areas p k , the connection point is calculated using the following formula p k RCS value p kRCS : ; in, p jr Radar trace p j The number of distance dimensions, Radar trace p j The number of angle dimensions, p jv Radar trace p j The number of velocity dimensions, p jRCS Radar trace p j of RCS value, p kr For connection point p k The number of distance dimensions, For connection point p k The number of angle dimensions, is the floor function.

6. The point trace aggregation method for extended targets of millimeter wave radar based on morphology according to claim 1, characterized in that: In S5, the points are condensed and the condensation center point and condensation area are determined, including: S51, use the following formula to calculate the center point of condensation p C The number of distance dimensions p Cr , number of angle dimensions , the number of speed dimensions p Cv and RCS value p CRCS : ; ; ; ; in, p l are the points participating in point condensation, that is, the points retained in the radar point visible graph after the corrosion operation. p lr Points participating in the aggregation of points p l The number of distance dimensions, Points participating in the aggregation of points p l The number of angle dimensions, p lv Points participating in the aggregation of points p l The number of velocity dimensions, p lRCS Points participating in the aggregation of points p l of RCS value, , L is the number of points participating in the point condensation; S52, determining the edge of the condensed area according to the condensed area edge judgment rule, thereby obtaining the condensed area; The condensed area edge judgment rules include: If the side outside the corresponding position of the point trace participating in the point trace condensation in the radar point trace visible graph is an isolation point or a blank area, then the side is the edge of the condensation area.