Dynamic target labeling method and related device

Abstract

The invention discloses a dynamic target marking method and a related device, and the method comprises the steps: carrying out the timestamp alignment of a millimeter-wave radar data set and a camera image data set, which are obtained in a preset time period, and obtaining a target millimeter-wave radar data set and a target camera image data set; screening potential millimeter wave radar dynamic point cloud data from the point attribute data set, and projecting the potential millimeter wave radar dynamic point cloud data and non-potential millimeter wave radar dynamic point cloud data to a point cloud interface to obtain point cloud interface data; converting the point cloud position data set and the potential millimeter wave radar dynamic point cloud data into a camera coordinate system, fusing with a target camera image data set, and projecting to an image interface to obtain image interface data; and determining each potential millimeter wave radar dynamic point of which the position is changed in the point cloud interface data and the image interface data by adopting an inter-frame switching function, and marking the potential millimeter wave radar dynamic points. According to the method and the device, the dynamic target can be accurately calibrated by using the time sequence change of the multi-frame point cloud data and the principle that the appearance characteristics of the dynamic target can be identified by the image.

Description

Technical Field

[0001] The present invention relates to the technical field of data annotation, and more specifically, to a dynamic target annotation method and related devices. Background Art

[0002] In the field of autonomous driving, millimeter-wave radar is widely used in target detection and tracking due to its excellent anti-interference ability and all-weather working characteristics. Existing millimeter-wave radars mainly perform target annotation based on single-frame point cloud data. The resolution of single-frame point cloud data generated by current millimeter-wave radars is low. Affected by noise and missing information, traditional single-frame point cloud data easily ignores the target motion trajectory, making it difficult to annotate dynamic targets in dynamic scenes, thus affecting the intelligent perception effect of millimeter-wave radar in applications such as autonomous driving. Summary of the invention

[0003] In view of this, the present invention discloses a dynamic target labeling method and related devices to achieve the labeling of dynamic targets in dynamic scenes and improve the intelligent perception effect of millimeter wave radar in applications such as autonomous driving.

[0004] A dynamic target labeling method, comprising:

[0005] Performing timestamp alignment on the millimeter-wave radar dataset and the camera image dataset acquired within a preset time period to obtain a target millimeter-wave radar dataset and a target camera image dataset, wherein the target millimeter-wave radar dataset includes: a point cloud position dataset and a point attribute dataset;

[0006] Screening out potential millimeter-wave radar dynamic point cloud data from the point attribute data set;

[0007] Projecting the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set onto a point cloud interface to obtain point cloud interface data;

[0008] The point cloud position data set and the potential millimeter wave radar dynamic point cloud data are converted into a camera coordinate system, and are fused with the target camera image data set and then projected onto an image interface to obtain image interface data;

[0009] An inter-frame switching function is used to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, and each potential millimeter-wave radar dynamic point is marked as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked.

[0010] Optionally, the performing timestamp alignment on the millimeter-wave radar dataset and the camera image dataset acquired within a preset time period to obtain a target millimeter-wave radar dataset and a target camera image dataset includes:

[0011] Within the preset time period, each frame of millimeter-wave radar data with a timestamp is obtained from the millimeter-wave radar and is sequentially put into a millimeter-wave radar data queue to obtain the millimeter-wave radar data set;

[0012] Within the preset time period, each frame of camera image data with a timestamp is acquired from the camera sensor and sequentially placed into a camera image data queue to obtain the camera image data set;

[0013] Reading each frame of millimeter-wave radar data in the millimeter-wave radar data set and each frame of camera image data in the camera image data set in sequence;

[0014] For the timestamp of each frame of millimeter-wave radar data, a linear interpolation method is used to determine the target frame camera image data with the corresponding timestamp from all the camera image data frames, so as to obtain the target millimeter-wave radar data set and the target camera image data set with aligned timestamps.

[0015] Optionally, screening out potential millimeter-wave radar dynamic point cloud data from the point attribute data set includes:

[0016] Read each frame of point attribute data in the point attribute data set in sequence;

[0017] Determine the Euclidean distance and speed of each frame point attribute data;

[0018] Filtering out all target frame point attribute data whose Euclidean distance is not greater than a distance threshold and whose speed is not less than a speed threshold from the point attribute data set;

[0019] The millimeter-wave radar points corresponding to the attribute data of each target frame point constitute the potential millimeter-wave radar dynamic point cloud data.

[0020] Optionally, the step of converting the point cloud position data set and the potential millimeter wave radar dynamic point cloud data into a camera coordinate system, fusing them with the target camera image data set and projecting them onto an image interface to obtain image interface data includes:

[0021] According to the camera intrinsic parameter matrix and the extrinsic parameter matrix from the millimeter-wave radar coordinate system to the camera coordinate system, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are converted into the camera coordinate system to obtain the target point cloud position data set and the target potential millimeter-wave radar dynamic point cloud data;

[0022] In the camera coordinate system, the target point cloud position data set, the target potential millimeter wave radar dynamic point cloud data and the target camera image data set are fused to obtain fused data;

[0023] The fused data is projected onto the image interface to obtain the image interface data.

[0024] Optionally, the inter-frame switching function is used to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, and each potential millimeter-wave radar dynamic point is marked as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked, including:

[0025] Adopting an inter-frame switching function to dynamically switch each frame of the point cloud interface data, determining potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of the point cloud interface data, and obtaining a first millimeter-wave radar dynamic point set;

[0026] Adopting an inter-frame switching function to dynamically switch each frame of the image interface data, determining potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of the image interface data, and obtaining a second millimeter-wave radar dynamic point set;

[0027] Determine each identical potential millimeter-wave radar dynamic point in the first millimeter-wave radar dynamic point set and the second millimeter-wave radar dynamic point set;

[0028] Each of the same potential millimeter-wave radar dynamic points is marked as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked.

[0029] Optionally, the dynamically switching each frame of the point cloud interface data in the point cloud interface data by using an inter-frame switching function includes:

[0030] When a forward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the previous frame of point cloud interface data;

[0031] When a backward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the next frame of point cloud interface data.

[0032] Optionally, the dynamically switching each frame of the image interface data in the image interface data by using an inter-frame switching function includes:

[0033] When a forward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the previous frame of image interface data;

[0034] When the backward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the next frame of image interface data.

[0035] A dynamic target marking device, comprising:

[0036] A timestamp alignment unit is used to perform timestamp alignment on a millimeter-wave radar data set and a camera image data set acquired within a preset time period to obtain a target millimeter-wave radar data set and a target camera image data set, wherein the target millimeter-wave radar data set includes: a point cloud position data set and a point attribute data set;

[0037] A screening unit, used for screening out potential millimeter wave radar dynamic point cloud data from the point attribute data set;

[0038] A first projection unit is used to project the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set onto a point cloud interface to obtain point cloud interface data;

[0039] A second projection unit is used to convert the point cloud position data set and the potential millimeter wave radar dynamic point cloud data into a camera coordinate system, and fuse them with the target camera image data set and then project them onto an image interface to obtain image interface data;

[0040] The dynamic point marking unit is used to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data by using an inter-frame switching function, and mark each of the potential millimeter-wave radar dynamic points as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point marking.

[0041] A computer storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the above-mentioned dynamic target marking method is implemented.

[0042] An electronic device, comprising: a memory and a processor;

[0043] The memory is used to store at least one instruction;

[0044] The processor is used to execute the at least one instruction to implement the above-mentioned dynamic target marking method.

[0045] It can be seen from the above technical scheme that the present invention discloses a method and related devices for marking dynamic targets, which perform timestamp alignment on a millimeter-wave radar data set and a camera image data set acquired within a preset time period to obtain a target millimeter-wave radar data set and a target camera image data set, screen out potential millimeter-wave radar dynamic point cloud data from a point attribute data set, project both the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set to a point cloud interface to obtain point cloud interface data, convert both the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data into a camera coordinate system, and after being fused with the target camera image data set, project them to an image interface to obtain image interface data, use an inter-frame switching function to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, determine the target millimeter-wave radar dynamic point for marking, and thereby obtain the target millimeter-wave radar point cloud data after dynamic point marking. This application processes the millimeter-wave radar dataset and the camera image dataset as a whole, and uses the principle that the multi-frame point cloud data in the millimeter-wave radar dataset can capture the time series changes reflected by the dynamic changes of dynamic targets in space, and the images in the camera image dataset can recognize the appearance characteristics of dynamic targets. It can simultaneously process the fine features of static targets and the motion trajectories of dynamic targets, ensuring that accurate calibration of dynamic targets can be achieved even when the point cloud is sparse and information is missing or contains noise, thereby improving the intelligent perception effect of millimeter-wave radar in applications such as autonomous driving and better adapting to complex real-world scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] 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 description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the disclosed drawings without paying creative work.

[0047] Figure 1 A flow chart of a dynamic target labeling method disclosed in an embodiment of the present invention;

[0048] Figure 2 A flow chart of a method for aligning timestamps of a millimeter-wave radar dataset and a camera image dataset disclosed in an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of an image interface and a point cloud interface before annotation disclosed in an embodiment of the present invention;

[0050] Figure 4 A schematic diagram of a labeled image interface and a point cloud interface disclosed in an embodiment of the present invention;

[0051] Figure 5 A schematic diagram of the structure of a dynamic target labeling device disclosed in an embodiment of the present invention;

[0052] Figure 6 The present invention is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0054] The embodiment of the present invention discloses a method for labeling dynamic targets and related devices, which process a millimeter-wave radar data set and a camera image data set as a whole, and use the principle that multi-frame point cloud data in the millimeter-wave radar data set can capture the time series changes reflected by the dynamic changes of dynamic targets in space, and that images in the camera image data set can recognize the appearance characteristics of dynamic targets. The fine features of static targets and the motion trajectories of dynamic targets can be processed simultaneously, ensuring that accurate calibration of dynamic targets can be achieved even when the point cloud is sparse, resulting in missing information or containing noise, thereby improving the intelligent perception effect of millimeter-wave radar in applications such as autonomous driving and better adapting to complex real-world scenarios.

[0055] See also Figure 1 , a flow chart of a dynamic target labeling method disclosed in an embodiment of the present invention, the method comprising:

[0056] Step S101: aligning timestamps of a millimeter-wave radar dataset and a camera image dataset acquired within a preset time period to obtain a target millimeter-wave radar dataset and a target camera image dataset.

[0057] It should be noted that the millimeter wave radar data set and camera image data set obtained in this application are data obtained within the same preset time period. The value of the preset time period is determined according to actual needs, for example, 1 hour, and this application does not limit it here.

[0058] In practical applications, the millimeter-wave radar dataset is obtained by a millimeter-wave radar installed on a vehicle, and the camera image dataset is obtained by a camera sensor installed on the same vehicle.

[0059] Since the working principles and sampling frequencies of millimeter-wave radar and camera sensors are usually different, the data they obtain may differ in time points. In order to ensure that the millimeter-wave radar dataset and the camera image dataset are synchronized in time so that they can be more accurately associated and fused later, this application aligns the timestamps of the millimeter-wave radar dataset and the camera image dataset, and adjusts the data at different time points to the same time base, thereby ensuring data synchronization.

[0060] Among them, the target millimeter wave radar dataset includes: point cloud position dataset and point attribute dataset.

[0061] The point cloud location dataset refers to the location information of the target object detected by the millimeter-wave radar in three-dimensional space.

[0062] Point attribute data set refers to the specific attributes and parameters related to each detection point (or target point, reflection point) detected and acquired by the millimeter wave radar system, such as ID number, lateral distance, longitudinal distance, lateral speed, longitudinal speed, label, angular velocity, length, width, etc.

[0063] Step S102: Filter out potential millimeter-wave radar dynamic point cloud data from the point attribute data set.

[0064] The potential millimeter-wave radar dynamic point cloud data in this application refers to data that has a high probability of becoming a target millimeter-wave radar dynamic point.

[0065] Step S103: Project the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set onto the point cloud interface to obtain point cloud interface data.

[0066] After the potential millimeter-wave radar dynamic point cloud data is screened out from the point attribute data set, the remaining data is the non-potential millimeter-wave radar dynamic point cloud data, which is actually static point cloud data.

[0067] Step S104: convert the point cloud position data set and the potential millimeter wave radar dynamic point cloud data into the camera coordinate system, fuse them with the target camera image data set, and project them onto the image interface to obtain image interface data.

[0068] In this application, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are both in the millimeter-wave radar coordinate system, and the target camera image data set is in the camera coordinate system. Therefore, before data fusion, the data needs to be converted to the same coordinate system. In this embodiment, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are converted to the camera coordinate system, and then fused with the target camera image data set in the camera coordinate system to obtain fused data, and the image interface data is obtained by projecting the fused data onto the image interface.

[0069] The point cloud data displayed by the point cloud interface in this application contains position information in three-dimensional space. Therefore, the point cloud interface provides the display and interaction of three-dimensional information. Through the point cloud interface, the shape, structure and position information of the object can be intuitively seen.

[0070] The image data displayed on the image interface is two-dimensional, so the image interface provides the display and interaction of two-dimensional information. Through the image interface, you can view the color, texture, shape and other features in the image.

[0071] Step S105, using the inter-frame switching function to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, and marking each of the potential millimeter-wave radar dynamic points as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked.

[0072] In practical applications, this application uses the inter-frame switching function, and utilizes the time series changes reflected by the multi-frame point cloud in the point cloud interface data, as well as the high-resolution details such as image texture provided in the image interface data, to achieve accurate labeling of the target millimeter-wave radar dynamic points. Therefore, by fusing multi-frame point cloud data and image information, this application can simultaneously process the fine features of static objects and the motion trajectory of dynamic targets, improve the point cloud labeling effect of millimeter-wave radar in dynamic environments, and enhance the robustness and reliability of the system in dynamic target detection and tracking. And it can effectively label whether the target is stationary or moving, so as to improve the detection and labeling capabilities of millimeter-wave radar point clouds in complex dynamic scenes.

[0073] In summary, the present invention discloses a method for marking dynamic targets, which performs timestamp alignment on a millimeter-wave radar data set and a camera image data set acquired within a preset time period to obtain a target millimeter-wave radar data set and a target camera image data set, screens out potential millimeter-wave radar dynamic point cloud data from a point attribute data set, projects both the potential millimeter-wave radar dynamic point cloud data and non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set to a point cloud interface to obtain point cloud interface data, converts both the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data into a camera coordinate system, and after being fused with the target camera image data set, projects them to an image interface to obtain image interface data, uses an inter-frame switching function to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, determines the target millimeter-wave radar dynamic points for marking, and thereby obtains the target millimeter-wave radar point cloud data after dynamic point marking. This application processes the millimeter-wave radar dataset and the camera image dataset as a whole, and uses the principle that the multi-frame point cloud data in the millimeter-wave radar dataset can capture the time series changes reflected by the dynamic changes of dynamic targets in space, and the images in the camera image dataset can recognize the appearance characteristics of dynamic targets. It can simultaneously process the fine features of static targets and the motion trajectories of dynamic targets, ensuring that accurate calibration of dynamic targets can be achieved even when the point cloud is sparse and information is missing or contains noise, thereby improving the intelligent perception effect of millimeter-wave radar in applications such as autonomous driving and better adapting to complex real-world scenarios.

[0074] In addition, unlike the traditional method that only relies on single frame or point cloud data, this application can better handle moving targets in dynamic scenes, improve annotation accuracy and system robustness. This combined method is not only suitable for dynamic target detection, but also enhances the system's ability to handle complex scenes such as fast-moving objects and occlusions, and significantly improves the perception effect of millimeter-wave radar point clouds in applications such as autonomous driving.

[0075] In one embodiment, see Figure 2 , a flowchart of a method for aligning timestamps of a millimeter-wave radar dataset and a camera image dataset disclosed in an embodiment of the present invention, that is, step S101 may specifically include:

[0076] Step S201: within a preset time period, each frame of millimeter-wave radar data with a timestamp is obtained from the millimeter-wave radar and is sequentially put into a millimeter-wave radar data queue to obtain a millimeter-wave radar data set.

[0077] In actual applications, an empty millimeter-wave radar data queue and an empty camera image data queue can be created. The length of the queue is determined according to actual needs, for example, 20, which is not limited in this application.

[0078] By putting each frame of millimeter-wave radar data with a timestamp obtained from the millimeter-wave radar into the created millimeter-wave radar data queue, a millimeter-wave radar data set can be obtained.

[0079] Step S202: Within a preset time period, obtain each frame of camera image data with a timestamp from the camera sensor and sequentially put it into the camera image data queue to obtain a camera image data set.

[0080] By putting each frame of camera image data with a timestamp obtained from the camera sensor into the camera image data queue, a camera image data set can be obtained.

[0081] Step S203: Sequentially read each frame of millimeter-wave radar data in the millimeter-wave radar data set and each frame of camera image data in the camera image data set.

[0082] Among them, sequentially reading each frame of millimeter-wave radar data in the millimeter-wave radar data set is actually sequentially reading each frame of millimeter-wave radar data in the millimeter-wave radar data queue.

[0083] Similarly, sequentially reading each frame of camera image data in the camera image data set is actually sequentially reading each frame of camera image data in the camera image data queue.

[0084] Step S204: For the timestamp of each frame of millimeter-wave radar data, use the linear interpolation method to determine the target frame of camera image data with the corresponding timestamp from all the frames of camera image data, and obtain the target millimeter-wave radar data set and the target camera image data set with aligned timestamps.

[0085] Specifically, sequentially read each frame of millimeter-wave radar data in the millimeter-wave radar data queue, and record the timestamp of the current frame of millimeter-wave radar data as Tr. Then sequentially read each frame of camera image data in the camera image data queue, record the timestamp of the current frame of camera image data as Tc1, and record the timestamp of the next frame of camera image data as Tc2.

[0086] The process of using the linear interpolation method is as follows:

[0087] When Tc1 < Tr < Tc2, calculate the absolute value of the difference between Tc1 - Tr as a1, and calculate the absolute value of the difference between Tc2 - Tr as a2.

[0088] When a1 < a2, change the value of Tc1 to Tr;

[0089] When a1 > a2, change the value of Tc2 to Tr.

[0090] Obtain the camera image data corresponding to the millimeter-wave radar data with the timestamp Tr.

[0091] Save the millimeter wave radar data and camera image data with timestamp Tr.

[0092] According to the above timestamp alignment method, a target millimeter-wave radar data set and a target camera image data set with timestamp alignment are obtained, wherein the target millimeter-wave radar data set is a collection of multiple millimeter-wave radar data with timestamp alignment, and the target camera image data set is a collection of multiple camera image data with timestamp alignment.

[0093] It should be noted that each millimeter-wave radar data in the target millimeter-wave radar data set includes: point cloud position data and point attribute data, and the timestamps of the point cloud position data and the point attribute data are the same as the timestamps of the corresponding millimeter-wave radar data. For example, the x-axis data and y-axis data of all points in the millimeter-wave radar data with a timestamp of Tr constitute the point cloud position data of the millimeter-wave radar with a timestamp of Tr.

[0094] In one embodiment, step S102 may specifically include:

[0095] (1) Read each frame of point attribute data in the point attribute dataset in sequence.

[0096] (2) Determine the Euclidean distance and speed of each frame point attribute data.

[0097] (3) Filtering out all target frame point attribute data whose Euclidean distance is not greater than a distance threshold and whose speed is not less than a speed threshold from the point attribute data set;

[0098] (4) The millimeter-wave radar points corresponding to the target frame point attribute data are used to form the potential millimeter-wave radar dynamic point cloud data.

[0099] The value of the distance threshold is determined according to actual needs and is not limited in the present invention.

[0100] Specifically, each frame of point attribute data Pi in the point attribute data set is read in sequence, and the x-axis position data of each frame of point attribute data Pi is defined as Pix and the y-axis position data is defined as Piy.

[0101] The purpose of distance threshold filtering is to remove noise points that are far away from the millimeter-wave radar and reduce data redundancy.

[0102] Set the x-axis distance threshold thre_x and the y-axis distance threshold thre_y. If the x-axis position data Pix of the point attribute data Pi exceeds the distance threshold thre_x, and / or the y-axis position data Piy of the point attribute data Pi exceeds the distance threshold thre_y, discard the millimeter-wave radar points corresponding to the point attribute data that exceed the x-axis distance threshold thre_x and / or the y-axis distance threshold thre_y.

[0103] The principle of speed threshold filtering is: select points that may be dynamic according to the speed attribute as potential millimeter-wave radar dynamic points. Calculate the speed value Vp of the millimeter-wave radar point according to the speed of the x-axis direction and the speed of the y-axis direction of the millimeter-wave radar point, and compare it with the speed Vb of the previous frame of millimeter-wave radar point cloud as the speed threshold. When Vp exceeds Vb, the millimeter-wave radar point is considered to be a potential millimeter-wave radar dynamic point.

[0104] In one embodiment, step S104 may specifically include:

[0105] (1) According to the camera intrinsic parameter matrix and the extrinsic parameter matrix from the millimeter-wave radar coordinate system to the camera coordinate system, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are converted into the camera coordinate system to obtain the target point cloud position data set and the target potential millimeter-wave radar dynamic point cloud data.

[0106] (2) In the camera coordinate system, the target point cloud position data set, the target potential millimeter wave radar dynamic point cloud data, and the target camera image data set are fused to obtain fused data.

[0107] (3) Projecting the fused data onto the image interface to obtain the image interface data.

[0108] Specifically, each frame of millimeter-wave radar point cloud position data in the point cloud position data set, each frame of potential millimeter-wave radar dynamic point in the potential millimeter-wave radar dynamic point cloud data, and each frame of target camera image data in the target camera image data set are read in sequence.

[0109] The millimeter-wave radar point cloud position data of the current frame is Ci, the potential millimeter-wave radar dynamic point of the current frame is Fdi, and the camera image data of the current frame is Ii.

[0110] According to the camera intrinsic parameter matrix Ri and the extrinsic parameter matrix Te from the millimeter-wave radar coordinate system to the camera coordinate system, Ci and Fdi are converted to the camera coordinate system to obtain the target point cloud position data set and the target potential millimeter-wave radar dynamic point cloud data.

[0111] In the camera coordinate system, the target point cloud position data set, the target potential millimeter-wave radar dynamic point cloud data and the target camera image data set are fused to obtain fused data, and the fused data is projected onto the image interface to obtain image interface data.

[0112] The potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set are projected onto the point cloud interface to obtain the point cloud interface data.

[0113] See also Figure 3 The schematic diagram of the image interface and the point cloud interface is shown. In the image interface data displayed in the image interface, black represents static points and green represents the screened potential millimeter-wave radar dynamic points.

[0114] Similarly, in the point cloud interface data displayed in the point cloud interface, black represents static points and green represents the screened potential millimeter-wave radar dynamic points.

[0115] In practical applications, a rectangular box can be drawn for each potential millimeter-wave radar dynamic point on the image interface to visually identify these potential millimeter-wave radar dynamic points.

[0116] In one embodiment, step S105 may specifically include:

[0117] (1) Using an inter-frame switching function to dynamically switch each frame of the point cloud interface data, determine potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of the point cloud interface data, and obtain a first millimeter-wave radar dynamic point set.

[0118] (2) Using an inter-frame switching function to dynamically switch each frame of the image interface data, determine potential millimeter-wave radar dynamic points whose positions change in adjacent frames of the image interface data, and obtain a second millimeter-wave radar dynamic point set.

[0119] (3) Determine each identical potential millimeter-wave radar dynamic point in the first millimeter-wave radar dynamic point set and the second millimeter-wave radar dynamic point set.

[0120] (4) Marking each of the same potential millimeter-wave radar dynamic points as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked.

[0121] The inter-frame switching function refers to switching between adjacent frames, for example, switching from a current frame to a previous frame, or from a current frame to a next frame.

[0122] This application can greatly improve the accuracy of dynamic target calibration by comparing the first millimeter wave radar dynamic point set determined based on the point cloud interface data with the second millimeter wave radar dynamic point set determined based on the image interface data, and taking the same potential millimeter wave radar dynamic points in the two millimeter wave radar dynamic point sets as the final target millimeter wave radar dynamic points. This can improve the intelligent perception effect of millimeter wave radar in applications such as autonomous driving and better adapt to complex real scenes.

[0123] In one embodiment, the process of dynamically switching each frame of point cloud interface data in the point cloud interface data using the inter-frame switching function may include:

[0124] When a forward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the previous frame of point cloud interface data;

[0125] When a backward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the next frame of point cloud interface data.

[0126] Assuming that the user generates a forward visualization data signal through the "s" key of the touch keyboard, after the "s" key of the touch keyboard generates the forward visualization data signal, the current frame point cloud interface data in the point cloud interface data can be switched to the previous frame point cloud interface data.

[0127] The backward visualization data signal is generated by the "d" key of the touch keyboard. After the backward visualization data signal is generated by the "d" key of the touch keyboard, the current frame point cloud interface data in the point cloud interface data can be switched to the next frame point cloud interface data.

[0128] In practical applications, the "s" and "d" keys on the keyboard can be used to switch between the previous frame of point cloud interface data and the next frame of point cloud interface data to achieve switching between multiple frames of point cloud interface data. In this way, potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of point cloud interface data can be determined to obtain the first millimeter-wave radar dynamic point set.

[0129] In one embodiment, the process of dynamically switching each frame of the image interface data using the inter-frame switching function may include:

[0130] When a forward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the previous frame of image interface data;

[0131] When the backward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the next frame of image interface data.

[0132] Similarly, assuming that the user generates a forward visualization data signal through the "s" key of the touch keyboard, after the "s" key of the touch keyboard generates the forward visualization data signal, the current frame image interface data in the image interface data can be switched to the previous frame image interface data.

[0133] The backward visualization data signal is generated by the "d" key of the touch keyboard. After the "d" key of the touch keyboard generates the backward visualization data signal, the current frame image interface data in the image interface data can be switched to the next frame image interface data.

[0134] In practical applications, the "s" and "d" keys on the keyboard can be used to switch between the previous frame image interface data and the next frame image interface data to achieve switching between multiple frames of image interface data. In this way, potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of image interface data can be determined to obtain a second millimeter-wave radar dynamic point set.

[0135] To facilitate user viewing, each identical potential millimeter-wave radar dynamic point in the first millimeter-wave radar dynamic point set and the second millimeter-wave radar dynamic point set may be marked.

[0136] See also Figure 4 As shown, the labeling process can be as follows:

[0137] Select and mark: Use "Shift+left mouse button" to select a point in the point cloud interface or image interface. The selected point is displayed in red, indicating that it is marked as a target millimeter-wave radar dynamic point.

[0138] Cancel marking: Use "Shift+left mouse button" to click the selected red point, and the point will be displayed in black, indicating that the marking is canceled.

[0139] After all target millimeter-wave radar dynamic points are marked, the position data of all points in the point cloud interface are read, and the attribute data marked as target millimeter-wave radar dynamic points and the attribute data not marked as target millimeter-wave radar dynamic points are saved as files. The class attribute of each point in the file indicates whether it is marked as a dynamic point.

[0140] Class attribute description:

[0141] class=1: The point is marked as a target millimeter-wave radar dynamic point.

[0142] class=0: The point is not marked, indicating that it is a static point.

[0143] This completes the marking of millimeter-wave radar dynamic points.

[0144] Corresponding to the above method embodiment, the present invention also discloses a dynamic target labeling device.

[0145] See also Figure 5 , a schematic diagram of a structure of a dynamic target labeling device disclosed in an embodiment of the present invention, the device may include:

[0146] The timestamp alignment unit 301 is used to perform timestamp alignment on the millimeter-wave radar data set and the camera image data set acquired within a preset time period to obtain a target millimeter-wave radar data set and a target camera image data set.

[0147] It should be noted that the millimeter wave radar data set and camera image data set obtained in this application are data obtained within the same preset time period. The value of the preset time period is determined according to actual needs, for example, 1 hour, and this application does not limit it here.

[0148] In practical applications, the millimeter-wave radar dataset is obtained by a millimeter-wave radar installed on a vehicle, and the camera image dataset is obtained by a camera sensor installed on the same vehicle.

[0149] Since the working principles and sampling frequencies of millimeter-wave radar and camera sensors are usually different, the data they obtain may differ in time points. In order to ensure that the millimeter-wave radar dataset and the camera image dataset are synchronized in time so that they can be more accurately associated and fused later, this application aligns the timestamps of the millimeter-wave radar dataset and the camera image dataset, and adjusts the data at different time points to the same time base, thereby ensuring data synchronization.

[0150] Among them, the target millimeter wave radar dataset includes: point cloud position dataset and point attribute dataset.

[0151] The point cloud location dataset refers to the location information of the target object detected by the millimeter-wave radar in three-dimensional space.

[0152] Point attribute data set refers to the specific attributes and parameters related to each detection point (or target point, reflection point) detected and acquired by the millimeter wave radar system, such as ID number, lateral distance, longitudinal distance, lateral speed, longitudinal speed, label, angular velocity, length, width, etc.

[0153] The screening unit 302 is used to screen out potential millimeter-wave radar dynamic point cloud data from the point attribute data set.

[0154] The potential millimeter-wave radar dynamic point cloud data in this application refers to data that has a high probability of becoming a target millimeter-wave radar dynamic point.

[0155] The first projection unit 303 is used to project the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set onto a point cloud interface to obtain point cloud interface data.

[0156] After the potential millimeter-wave radar dynamic point cloud data is screened out from the point attribute data set, the remaining data is the non-potential millimeter-wave radar dynamic point cloud data, which is actually static point cloud data.

[0157] The second projection unit 304 is used to convert the point cloud position data set and the potential millimeter wave radar dynamic point cloud data into a camera coordinate system, and fuse them with the target camera image data set and then project them onto an image interface to obtain image interface data.

[0158] In this application, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are both in the millimeter-wave radar coordinate system, and the target camera image data set is in the camera coordinate system. Therefore, before data fusion, the data needs to be converted to the same coordinate system. In this embodiment, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are converted to the camera coordinate system, and then fused with the target camera image data set in the camera coordinate system to obtain fused data, and the image interface data is obtained by projecting the fused data onto the image interface.

[0159] The point cloud data displayed by the point cloud interface in this application contains position information in three-dimensional space. Therefore, the point cloud interface provides the display and interaction of three-dimensional information. Through the point cloud interface, the shape, structure and position information of the object can be intuitively seen.

[0160] The image data displayed on the image interface is two-dimensional, so the image interface provides the display and interaction of two-dimensional information. Through the image interface, you can view the color, texture, shape and other features in the image.

[0161] The dynamic point marking unit 305 is used to use the inter-frame switching function to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, and mark each of the potential millimeter-wave radar dynamic points as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point marking.

[0162] In practical applications, this application uses the inter-frame switching function, and utilizes the time series changes reflected by the multi-frame point cloud in the point cloud interface data, as well as the high-resolution details such as image texture provided in the image interface data, to achieve accurate labeling of the target millimeter-wave radar dynamic points. Therefore, by fusing multi-frame point cloud data and image information, this application can simultaneously process the fine features of static objects and the motion trajectory of dynamic targets, improve the point cloud labeling effect of millimeter-wave radar in dynamic environments, and enhance the robustness and reliability of the system in dynamic target detection and tracking. And it can effectively label whether the target is stationary or moving, so as to improve the detection and labeling capabilities of millimeter-wave radar point clouds in complex dynamic scenes.

[0163] In summary, the present invention discloses a dynamic target annotation device, which performs timestamp alignment on a millimeter-wave radar data set and a camera image data set acquired within a preset time period to obtain a target millimeter-wave radar data set and a target camera image data set, screens out potential millimeter-wave radar dynamic point cloud data from a point attribute data set, projects both the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set to a point cloud interface to obtain point cloud interface data, converts both the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data into a camera coordinate system, and after being fused with the target camera image data set, projects them to an image interface to obtain image interface data, uses an inter-frame switching function to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, determines the target millimeter-wave radar dynamic points for annotation, and thereby obtains the target millimeter-wave radar point cloud data after dynamic point annotation. This application processes the millimeter-wave radar dataset and the camera image dataset as a whole, and uses the principle that the multi-frame point cloud data in the millimeter-wave radar dataset can capture the time series changes reflected by the dynamic changes of dynamic targets in space, and the images in the camera image dataset can recognize the appearance characteristics of dynamic targets. It can simultaneously process the fine features of static targets and the motion trajectories of dynamic targets, ensuring that accurate calibration of dynamic targets can be achieved even when the point cloud is sparse and information is missing or contains noise, thereby improving the intelligent perception effect of millimeter-wave radar in applications such as autonomous driving and better adapting to complex real-world scenarios.

[0164] In addition, unlike the traditional method that only relies on single frame or point cloud data, this application can better handle moving targets in dynamic scenes, improve annotation accuracy and system robustness. This combined method is not only suitable for dynamic target detection, but also enhances the system's ability to handle complex scenes such as fast-moving objects and occlusions, and significantly improves the perception effect of millimeter-wave radar point clouds in applications such as autonomous driving.

[0165] In one embodiment, the timestamp alignment unit 301 may be specifically used for:

[0166] Within the preset time period, each frame of millimeter-wave radar data with a timestamp is obtained from the millimeter-wave radar and is sequentially put into a millimeter-wave radar data queue to obtain the millimeter-wave radar data set;

[0167] Within the preset time period, each frame of camera image data with a timestamp is acquired from the camera sensor and sequentially placed into a camera image data queue to obtain the camera image data set;

[0168] Reading each frame of millimeter-wave radar data in the millimeter-wave radar data set and each frame of camera image data in the camera image data set in sequence;

[0169] For the timestamp of each frame of millimeter-wave radar data, a linear interpolation method is used to determine the target frame camera image data with the corresponding timestamp from all the camera image data frames, so as to obtain the target millimeter-wave radar data set and the target camera image data set with aligned timestamps.

[0170] In one embodiment, the screening unit 302 may be specifically used to:

[0171] Read each frame of point attribute data in the point attribute data set in sequence;

[0172] Determine the Euclidean distance and speed of each frame point attribute data;

[0173] Filtering out all target frame point attribute data whose Euclidean distance is not greater than a distance threshold and whose speed is not less than a speed threshold from the point attribute data set;

[0174] The millimeter-wave radar points corresponding to the attribute data of each target frame point constitute the potential millimeter-wave radar dynamic point cloud data.

[0175] In one embodiment, the second projection unit 304 may be specifically used for:

[0176] According to the camera intrinsic parameter matrix and the extrinsic parameter matrix from the millimeter-wave radar coordinate system to the camera coordinate system, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are converted into the camera coordinate system to obtain the target point cloud position data set and the target potential millimeter-wave radar dynamic point cloud data;

[0177] In the camera coordinate system, the target point cloud position data set, the target potential millimeter wave radar dynamic point cloud data and the target camera image data set are fused to obtain fused data;

[0178] The fused data is projected onto the image interface to obtain the image interface data.

[0179] In one embodiment, the dynamic point labeling unit 305 may be specifically used for:

[0180] Adopting an inter-frame switching function to dynamically switch each frame of the point cloud interface data, determining potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of the point cloud interface data, and obtaining a first millimeter-wave radar dynamic point set;

[0181] Adopting an inter-frame switching function to dynamically switch each frame of the image interface data, determining potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of the image interface data, and obtaining a second millimeter-wave radar dynamic point set;

[0182] Determine each identical potential millimeter-wave radar dynamic point in the first millimeter-wave radar dynamic point set and the second millimeter-wave radar dynamic point set;

[0183] Each of the same potential millimeter-wave radar dynamic points is marked as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked.

[0184] In one embodiment, the dynamic point labeling unit 305 may be specifically used for:

[0185] When a forward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the previous frame of point cloud interface data;

[0186] When a backward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the next frame of point cloud interface data.

[0187] In one embodiment, the dynamic point labeling unit 305 may be specifically used for:

[0188] When a forward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the previous frame of image interface data;

[0189] When the backward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the next frame of image interface data.

[0190] It should be noted that, for the specific working principles of each component in the device embodiment, please refer to the corresponding part of the method embodiment, which will not be repeated here.

[0191] Corresponding to the above-mentioned embodiment, the present invention further discloses a computer storage medium, which stores at least one instruction, and when the at least one instruction is executed by a processor, the above-mentioned dynamic target marking method is implemented.

[0192] Computer storage media may be tangible media that may contain or store programs for use by or in conjunction with an instruction execution system, device, or apparatus. Computer storage media may be machine-readable signal media or machine-readable storage media. Computer storage media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0193] Corresponding to the above embodiment, Figure 6 As shown, the present invention also provides an electronic device, which may include: a processor 1 and a memory 2;

[0194] The processor 1 and the memory 2 communicate with each other via a communication bus 3;

[0195] Processor 1, configured to execute at least one instruction;

[0196] Memory 2, used to store at least one instruction;

[0197] The processor 1 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0198] The memory 2 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0199] Among them, the processor executes at least one instruction to implement the steps shown in the embodiment of the dynamic target marking method.

[0200] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0201] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0202] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for labeling a dynamic target, characterized in that: include: Performing timestamp alignment on the millimeter-wave radar dataset and the camera image dataset acquired within a preset time period to obtain a target millimeter-wave radar dataset and a target camera image dataset, wherein the target millimeter-wave radar dataset includes: a point cloud position dataset and a point attribute dataset; Screening out potential millimeter-wave radar dynamic point cloud data from the point attribute data set; Projecting the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set onto a point cloud interface to obtain point cloud interface data; The point cloud position data set and the potential millimeter wave radar dynamic point cloud data are converted into a camera coordinate system, and are fused with the target camera image data set and then projected onto an image interface to obtain image interface data; An inter-frame switching function is used to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, and each potential millimeter-wave radar dynamic point is marked as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked.

2. The dynamic target labeling method according to claim 1, characterized in that: The time stamp alignment of the millimeter wave radar data set and the camera image data set acquired within a preset time period to obtain a target millimeter wave radar data set and a target camera image data set includes: Within the preset time period, each frame of millimeter-wave radar data with a timestamp is obtained from the millimeter-wave radar and is sequentially put into a millimeter-wave radar data queue to obtain the millimeter-wave radar data set; Within the preset time period, each frame of camera image data with a timestamp is acquired from the camera sensor and sequentially placed into a camera image data queue to obtain the camera image data set; Reading each frame of millimeter-wave radar data in the millimeter-wave radar data set and each frame of camera image data in the camera image data set in sequence; For the timestamp of each frame of millimeter-wave radar data, a linear interpolation method is used to determine the target frame camera image data with the corresponding timestamp from all the camera image data frames, so as to obtain the target millimeter-wave radar data set and the target camera image data set with aligned timestamps.

3. The dynamic target labeling method according to claim 1 or 2, characterized in that: The step of screening out potential millimeter wave radar dynamic point cloud data from the point attribute data set includes: Read each frame of point attribute data in the point attribute data set in sequence; Determine the Euclidean distance and speed of each frame point attribute data; Filtering out all target frame point attribute data whose Euclidean distance is not greater than a distance threshold and whose speed is not less than a speed threshold from the point attribute data set; The millimeter-wave radar points corresponding to the attribute data of each target frame point constitute the potential millimeter-wave radar dynamic point cloud data.

4. The dynamic target labeling method according to claim 1 or 2, characterized in that: The step of converting the point cloud position data set and the potential millimeter wave radar dynamic point cloud data into a camera coordinate system, fusing them with the target camera image data set and projecting them onto an image interface to obtain image interface data includes: According to the camera intrinsic parameter matrix and the extrinsic parameter matrix from the millimeter-wave radar coordinate system to the camera coordinate system, the point cloud position data set and the potential millimeter-wave radar dynamic point cloud data are converted into the camera coordinate system to obtain the target point cloud position data set and the target potential millimeter-wave radar dynamic point cloud data; In the camera coordinate system, the target point cloud position data set, the target potential millimeter wave radar dynamic point cloud data and the target camera image data set are fused to obtain fused data; The fused data is projected onto the image interface to obtain the image interface data.

5. The dynamic target labeling method according to claim 1, characterized in that: The inter-frame switching function is used to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data, and each potential millimeter-wave radar dynamic point is marked as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked, including: Adopting an inter-frame switching function to dynamically switch each frame of the point cloud interface data, determining potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of the point cloud interface data, and obtaining a first millimeter-wave radar dynamic point set; Adopting an inter-frame switching function to dynamically switch each frame of the image interface data, determining potential millimeter-wave radar dynamic points whose positions have changed in adjacent frames of the image interface data, and obtaining a second millimeter-wave radar dynamic point set; Determine each identical potential millimeter-wave radar dynamic point in the first millimeter-wave radar dynamic point set and the second millimeter-wave radar dynamic point set; Each of the same potential millimeter-wave radar dynamic points is marked as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point is marked.

6. The dynamic target labeling method according to claim 5, characterized in that: The method of dynamically switching each frame of point cloud interface data in the point cloud interface data by using an inter-frame switching function includes: When a forward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the previous frame of point cloud interface data; When a backward visualization data signal is received, the current frame of point cloud interface data in the point cloud interface data is switched to the next frame of point cloud interface data.

7. The dynamic target labeling method according to claim 5 or 6, characterized in that: The dynamically switching each frame of the image interface data in the image interface data by using the inter-frame switching function includes: When a forward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the previous frame of image interface data; When the backward visualization data signal is received, the current frame of image interface data in the image interface data is switched to the next frame of image interface data.

8. A dynamic target labeling device, characterized in that: include: A timestamp alignment unit is used to perform timestamp alignment on a millimeter-wave radar data set and a camera image data set acquired within a preset time period to obtain a target millimeter-wave radar data set and a target camera image data set, wherein the target millimeter-wave radar data set includes: a point cloud position data set and a point attribute data set; A screening unit, used for screening out potential millimeter wave radar dynamic point cloud data from the point attribute data set; A first projection unit is used to project the potential millimeter-wave radar dynamic point cloud data and the non-potential millimeter-wave radar dynamic point cloud data in the point attribute data set onto a point cloud interface to obtain point cloud interface data; A second projection unit is used to convert the point cloud position data set and the potential millimeter wave radar dynamic point cloud data into a camera coordinate system, and fuse them with the target camera image data set and then project them onto an image interface to obtain image interface data; The dynamic point marking unit is used to determine each potential millimeter-wave radar dynamic point whose position has changed in the point cloud interface data and the image interface data by using an inter-frame switching function, and mark each of the potential millimeter-wave radar dynamic points as a target millimeter-wave radar dynamic point to obtain the target millimeter-wave radar point cloud data after the dynamic point marking.

9. A computer storage medium, characterized in that The computer storage medium stores at least one instruction, and when the at least one instruction is executed by the processor, the dynamic target labeling method as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: The electronic device comprises: a memory and a processor; The memory is used to store at least one instruction; The processor is used to execute the at least one instruction to implement the dynamic target labeling method as described in any one of claims 1 to 7.

Images