Point cloud processing method and device, terminal equipment and storage medium

By deleting points with vertical coordinates greater than the threshold and using plane fitting methods, the problem of low accuracy of point cloud detection in road surface during vehicle autonomous driving is solved, especially in the case of ups and downs, the removal of non-road surfaces is improved. point accuracy.

CN119091097APending Publication Date: 2024-12-06HAOMO TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310665238.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-06-06
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

When detecting point clouds on the road surface during vehicle autonomous driving, the prior art has the problem of removing non-road points with low accuracy, especially when the road surface has fluctuations or ups and downs.

Method used

By obtaining the point cloud to be processed, the points whose vertical coordinates are greater than the threshold are deleted, the pavement is fitted using the plane fitting method, and then the points whose distance from the fitted plane is greater than the threshold are deleted again to improve the accuracy of removing non-pavement points. Rate.

Benefits of technology

It effectively reduces the impact of road undulations and ups and downs on removing non-pavement points, improves the accuracy of removing non-pavement points, ensures that the fitting plane is closer to the real road surface, thereby improving the accuracy of detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091097A_ABST
    Figure CN119091097A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of data processing, and provides a point cloud processing method and device, terminal equipment and a storage medium. The method comprises the following steps: acquiring a to-be-processed first point cloud; deleting points of which the vertical direction coordinates are greater than a first threshold value in the first point cloud to obtain a second point cloud; fitting the second point cloud by using a plane fitting method to obtain a fitting plane of the road surface; and deleting points in the first point cloud, wherein the distance between the points and the fitting plane is greater than a second threshold value, and obtaining the processed first point cloud. Even for the road surface with fluctuating or uphill and downhill conditions, the fitting plane of the road surface obtained by removing most non-road surface points is closer to the real road surface, for example, the plane obtained by fitting the uphill and downhill road surfaces is also a plane with a gradient. Therefore, the non-pavement points in the first point cloud are filtered based on the fitting plane, and the accuracy of removing the non-pavement points can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a point cloud processing method, apparatus, terminal device and storage medium. Background Art

[0002] In the technical field of vehicle autonomous driving, it is usually necessary to accurately detect the point cloud data of the road surface, which involves the operation of removing non-road points, such as removing the point cloud of pedestrians and vehicles on the road. The existing technology usually detects non-road points based on the vertical coordinates of the point cloud. For example, the points in the point cloud with vertical coordinates greater than a set threshold can be deleted as non-road points. However, if the road surface is undulating or ups and downs, the accuracy of removing non-road points will be greatly reduced. Summary of the invention

[0003] In view of this, embodiments of the present application provide a point cloud processing method, apparatus, terminal device and storage medium, which can improve the accuracy of removing non-road surface points.

[0004] A first aspect of an embodiment of the present application provides a point cloud processing method, comprising:

[0005] Get the first point cloud to be processed;

[0006] Deleting points in the first point cloud whose vertical coordinates are greater than a first threshold value to obtain a second point cloud;

[0007] Fitting the second point cloud using a plane fitting method to obtain a fitting plane of the road surface;

[0008] Points in the first point cloud whose distances from the fitting plane are greater than a second threshold are deleted to obtain the processed first point cloud.

[0009] In an embodiment of the present application, the first point cloud to be processed is first obtained, and the points whose vertical coordinates in the first point cloud are greater than the first threshold are deleted to obtain the second point cloud; then, the second point cloud is fitted using a plane fitting method to obtain a fitting plane of the road surface; finally, the first point cloud is obtained again, and the points whose distances from the first point cloud to the fitting plane are greater than the second threshold are deleted to obtain the processed first point cloud. The above process can remove most of the non-road surface points by deleting the points whose vertical coordinates in the first point cloud are greater than the first threshold, thereby reducing interference with subsequent plane fitting. In actual situations, the road surface point cloud of a single-frame point cloud can be approximated as a plane, so the second point cloud obtained after removing most of the non-road surface points can be fitted into a fitting plane of the road surface. After obtaining the fitting plane, the non-road surface points in the first point cloud can be deleted by calculating the distance between each point in the first point cloud and the fitting plane. Even for roads with undulations or ups and downs, the fitted plane of the road surface obtained after removing most of the non-road surface points is closer to the real road surface. For example, the plane fitted for an ups and downs road surface is also a plane with a slope. Therefore, filtering the non-road surface points in the original point cloud (i.e., the first point cloud without removing the non-road surface points) based on the fitted plane can improve the accuracy of removing the non-road surface points.

[0010] In an implementation of the embodiment of the present application, the first point cloud is any one frame of point cloud in a continuous multi-frame point cloud, and the multi-frame point cloud is acquired by a radar device on a vehicle; the method may further include:

[0011] Obtaining each frame of point cloud after processing in the multi-frame point cloud by using the same processing method as the first point cloud;

[0012] Each frame of point cloud processed in the multiple frames of point cloud is spliced ​​to obtain a third point cloud.

[0013] Specifically, obtaining the first point cloud to be processed may include:

[0014] Acquire the first point cloud from the multi-frame point clouds;

[0015] The first point cloud is converted from a radar coordinate system to a vehicle body coordinate system corresponding to the first point cloud.

[0016] Furthermore, before converting the first point cloud from the radar coordinate system to the vehicle body coordinate system corresponding to the first point cloud, the method may further include:

[0017] Detecting a target distance between a scene position corresponding to the first point cloud and a scene position corresponding to a fourth point cloud; wherein the fourth point cloud is a previous frame point cloud adjacent to the first point cloud in the multi-frame point cloud;

[0018] If the target distance is greater than a third threshold, a step of converting the first point cloud from a radar coordinate system to a vehicle body coordinate system corresponding to the first point cloud and subsequent steps are performed.

[0019] Further, detecting the target distance between the scene position corresponding to the first point cloud and the scene position corresponding to the fourth point cloud may include:

[0020] Detecting the travel distance of the vehicle from a first time point to a second time point; wherein the first time point is the time point when the radar device collects the fourth point cloud, and the second time point is the time point when the radar device collects the first point cloud;

[0021] The travel distance is determined as the target distance.

[0022] Furthermore, after detecting the target distance between the scene position corresponding to the first point cloud and the scene position corresponding to the fourth point cloud, the method may further include:

[0023] If the target distance is less than or equal to the third threshold, the first point cloud is deleted from the multi-frame point cloud.

[0024] In an implementation of the embodiment of the present application, stitching each frame of the processed point cloud in the multiple frames of point cloud to obtain a third point cloud may include:

[0025] Each frame of point cloud processed in the multiple frames of point cloud is converted to the same coordinate system and point cloud merging processing is performed to obtain the third point cloud; wherein the same coordinate system is the vehicle body coordinate system corresponding to any frame of point cloud in the multiple frames of point cloud.

[0026] A second aspect of an embodiment of the present application provides a point cloud processing device, including:

[0027] A point cloud acquisition module, used for acquiring a first point cloud to be processed;

[0028] A first deletion module, used to delete points in the first point cloud whose vertical coordinates are greater than a first threshold, to obtain a second point cloud;

[0029] A plane fitting module, used for fitting the second point cloud using a plane fitting method to obtain a fitting plane of the road surface;

[0030] The second deletion module is used to delete points in the first point cloud whose distance from the fitting plane is greater than a second threshold, so as to obtain the processed first point cloud.

[0031] A third aspect of an embodiment of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the point cloud processing method provided in the first aspect of the embodiment of the present application is implemented.

[0032] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the point cloud processing method provided in the first aspect of the embodiment of the present application is implemented.

[0033] The fifth aspect of the embodiments of the present application provides a computer program product. When the computer program product is run on a terminal device, the terminal device executes the point cloud processing method provided in the first aspect of the embodiments of the present application.

[0034] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of a point cloud processing method provided in an embodiment of the present application;

[0036] Figure 2 It is a schematic diagram of the operation flow of the point cloud processing method provided by the embodiment of the present application in an actual application scenario;

[0037] Figure 3 is a structural schematic diagram of a point cloud processing device provided in an embodiment of the present application;

[0038] Figure 4 It is a schematic diagram of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures, technologies, etc. are proposed, so as to thoroughly understand the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details. In other cases, the detailed description of well-known systems, devices, circuits and methods is omitted to prevent unnecessary details from hindering the description of the present application. In addition, in the description of the present application specification and the attached claims, the terms "first", "second", "third" etc. are only used to distinguish the description, and cannot be interpreted as indicating or suggesting relative importance.

[0040] Nowadays, autonomous driving technology has developed rapidly. The various models run by the perception module of autonomous driving vehicles often require a large amount of data and true values ​​of real road scenes for training, but the current cost of obtaining data truth values ​​is high, and the efficiency of truth value annotation is low, which limits the upgrade iteration and optimization of the perception module. Among various types of perception functions, lane line detection and classification are particularly important. It is the basis for autonomous driving vehicles to achieve automatic cruising, lane change and overtaking, so the efficient acquisition of lane line truth values ​​has become an important issue. At present, 4D annotation is usually used to obtain lane line truth values. 4D annotation refers to splicing sensor data (such as point cloud data) of vehicles traveling for a period of time to obtain a global data, and then uniformly annotating targets such as lane lines and obstacles in the global data. The key technical point in 4D annotation is the splicing of road point clouds, and the difficulty of road point cloud splicing is that it is necessary not only to remove non-road points such as vehicles and pedestrians on the road, but also to consider the situation of vehicles going up and downhill and road undulations. The existing technology usually only detects non-road surface points based on the vertical coordinates of the point cloud. However, when the road surface is undulating or up and down, the accuracy of removing non-road surface points will be greatly reduced.

[0041] In view of this, the embodiments of the present application provide a point cloud processing method, apparatus, terminal device and storage medium, which can reduce the impact of road undulations and ups and downs, and improve the accuracy of removing non-road points. For more specific technical implementation details of the embodiments of the present application, please refer to the method embodiments described below.

[0042] It should be understood that the execution subjects of the various method embodiments of the present application are various types of terminal devices or servers, such as mobile phones, tablet computers, wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (PDA), large-screen TVs, etc. The embodiments of the present application do not impose any restrictions on the specific types of the terminal devices and servers.

[0043] See also Figure 1 , showing a point cloud processing method provided by an embodiment of the present application, comprising:

[0044] 101. Obtain a first point cloud to be processed;

[0045] First, the first point cloud to be processed is obtained, where the first point cloud refers to a frame of original point cloud without removing non-road points. In actual operation, the point cloud data of the road surface can be collected by a radar device (such as a laser radar) installed on the vehicle, and then the point cloud data is sent to the terminal device (i.e., the executor of the embodiment of the method), and the terminal device can obtain a frame of original point cloud from the received point cloud data as the first point cloud to be processed.

[0046] 102. Delete points whose vertical coordinates are greater than a first threshold in the first point cloud to obtain a second point cloud;

[0047] After obtaining the first point cloud, most of the points in the non-road area can be deleted according to the coordinates of the first point cloud in the vertical direction. Specifically, the points in the first point cloud whose vertical coordinates (i.e., z coordinates) are greater than the set threshold can be deleted, and then the first point cloud after the deletion of the points is determined as the second point cloud. For example, the threshold can be set to 20cm, and then the points in the first point cloud that are more than 20cm away from the z=0 plane can be deleted to obtain the second point cloud. Since the second point cloud is the point cloud obtained after deleting most of the non-road points, the second point cloud can also be called a road point cloud.

[0048] 103. Fit the second point cloud using a plane fitting method to obtain a fitting plane of the road surface;

[0049] After obtaining the second point cloud, various plane fitting methods can be used to fit the second point cloud, thereby obtaining a fitting plane of the road surface. In actual situations, the road surface point cloud of a single-frame point cloud can be approximated as a plane, so the road surface point cloud obtained in step 102 (i.e., the second point cloud) can be used to fit the plane equation of the road surface Ax+By+Cz+D=0 using various plane fitting methods, where A, B, C, and D are plane equation parameters; x, y, and z are point cloud coordinates. The plane represented by the plane equation Ax+By+Cz+D=0 is the fitting plane of the road surface, and the plane fitting methods used here may include but are not limited to: least squares method, RANSAC fitting method, and the like. In addition, since the fitting is to the road surface point cloud obtained after removing most of the non-road surface points, the fitting plane will be very close to the real road surface.

[0050] 104. Delete points in the first point cloud whose distances from the fitting plane are greater than a second threshold, to obtain a processed first point cloud.

[0051] After obtaining the fitting plane of the road surface, the non-road surface points in the first point cloud (i.e., the original point cloud without removing the non-road surface points) can be deleted based on the fitting plane. Specifically, the distance between each point in the first point cloud and the fitting plane can be calculated respectively, and then the points with a distance greater than a set threshold (e.g., 15 cm) are deleted as non-road surface points, thereby obtaining the processed first point cloud, that is, the first point cloud after deleting the non-road surface points. Among them, the distance Distance between any point (x, y, z) in the first point cloud and the fitting plane Ax+By+Cz+D=0 can be calculated by the following formula:

[0052]

[0053] Since the fitting plane is very close to the real road surface, the non-road surface points in the original point cloud can be deleted more accurately by calculating the distance between the point and the fitting plane.

[0054] In an implementation of the embodiment of the present application, the first point cloud is any one frame of point cloud in a continuous multi-frame point cloud, and the multi-frame point cloud is acquired by a radar device on a vehicle; the method may further include:

[0055] (1) obtaining each frame of point cloud after processing in the multi-frame point cloud by using the same processing method as the first point cloud;

[0056] (2) stitching each frame of point cloud after processing in the multiple frames of point cloud to obtain a third point cloud.

[0057] When the point cloud processing method provided in the embodiment of the present application is applied to the splicing of road point clouds, continuous multi-frame point clouds can be collected by a radar device (such as a laser radar) installed on a moving vehicle. For each frame of the multi-frame point cloud, non-road points can be removed in the same processing method as the first point cloud, thereby obtaining each frame of the processed point cloud in the multi-frame point cloud (that is, each frame of the point cloud after removing the non-road points). Then, each frame of the processed point cloud in the multi-frame point cloud is spliced ​​to obtain a spliced ​​point cloud (represented by a third point cloud).

[0058] In an implementation of the embodiment of the present application, stitching each frame of the processed point cloud in the multiple frames of point cloud to obtain a third point cloud may include:

[0059] Each frame of point cloud processed in the multiple frames of point cloud is converted to the same coordinate system and point cloud merging processing is performed to obtain the third point cloud; wherein the same coordinate system is the vehicle body coordinate system corresponding to any frame of point cloud in the multiple frames of point cloud.

[0060] When performing point cloud stitching, each frame of the processed point cloud in the multi-frame point cloud can be converted to the same coordinate system according to the position and posture of each frame of the point cloud, and the point cloud can be merged to finally obtain the third point cloud after stitching. The coordinate system converted here can be the vehicle body coordinate system corresponding to any frame of the point cloud in the multi-frame point cloud. For example, assuming that the multi-frame point cloud contains a frame of point cloud collected at time t1, then the point cloud corresponds to the vehicle body coordinate system at time t1. Each frame of the point cloud in the multi-frame point cloud can be converted to the vehicle body coordinate system, and the point cloud can be merged according to the position and posture of the point cloud, thereby completing the point cloud stitching process.

[0061] Specifically, obtaining the first point cloud to be processed may include:

[0062] (1) Acquire the first point cloud from the multi-frame point cloud;

[0063] (2) Converting the first point cloud from the radar coordinate system to the vehicle body coordinate system corresponding to the first point cloud.

[0064] The original point cloud obtained by the radar device is data in the radar coordinate system. The radar coordinate system is related to the position where it is installed on the vehicle. Therefore, the relationship between the radar coordinate system and the road surface is not fixed, which will cause inconvenience in subsequent point cloud computing. As for the vehicle's body coordinate system, it is relatively fixed, and its vertical direction (z coordinate axis) and the relationship with the road surface remain basically unchanged. Therefore, it is more convenient to convert the original point cloud to the body coordinate system for point cloud computing. Specifically, after obtaining a frame of original point cloud as the first point cloud from the multi-frame point cloud collected by the radar device, the first point cloud can be converted from the radar coordinate system to the body coordinate system corresponding to the first point cloud, so as to perform point cloud computing later.

[0065] Furthermore, before converting the first point cloud from the radar coordinate system to the vehicle body coordinate system corresponding to the first point cloud, the method may further include:

[0066] (1) detecting a target distance between a scene position corresponding to the first point cloud and a scene position corresponding to a fourth point cloud; wherein the fourth point cloud is a previous frame point cloud adjacent to the first point cloud in the multi-frame point cloud;

[0067] (2) If the target distance is greater than a third threshold, executing a step of converting the first point cloud from the radar coordinate system to a vehicle body coordinate system corresponding to the first point cloud and subsequent steps.

[0068] Generally speaking, the detection frequency of the radar device is about 10Hz. When the vehicle is traveling at a slow speed, the scene positions corresponding to two consecutive frames of point clouds collected by the radar device will be particularly close, which will lead to the splicing of a large amount of repeated point cloud data when splicing the point cloud, resulting in unnecessary calculation and data storage. To solve this problem, before processing the first point cloud, the distance between the scene position corresponding to the first point cloud and the scene position corresponding to the previous frame of point cloud adjacent to the first point cloud in the multi-frame point cloud can be detected, and the distance is represented by the target distance. If the target distance is greater than the set threshold, it means that the distance between the two adjacent frames of point clouds is not too close, and the first point cloud can be processed normally (converting the coordinate system and removing non-road points). On the contrary, if the target distance is less than or equal to the set threshold, it means that the distance between the two adjacent frames of point clouds is too close. At this time, the first point cloud can be deleted from the multi-frame point cloud, and then the next frame of point cloud can be read to continue the same judgment. Through this setting, a large amount of repeated point cloud data can be avoided when splicing point clouds, effectively reducing the calculation and storage of point cloud data.

[0069] Furthermore, detecting the target distance between the scene position corresponding to the first point cloud and the scene position corresponding to the fourth point cloud may include:

[0070] (1) detecting a travel distance of the vehicle from a first time point to a second time point; wherein the first time point is the time point at which the radar device collects the fourth point cloud, and the second time point is the time point at which the radar device collects the first point cloud;

[0071] (2) The travel distance is determined as the target distance.

[0072] When detecting the target distance between the scene positions corresponding to two adjacent point clouds, one feasible operation method is to detect the driving distance of the vehicle in the corresponding time. The two adjacent point clouds obtained here are the fourth point cloud and the first point cloud, respectively. Assuming that the time point when the radar device collects the fourth point cloud is t0 and the time point when the first point cloud is collected is t1, the driving distance of the vehicle in the time period from t0 to t1 can be detected and determined as the target distance.

[0073] In an embodiment of the present application, the first point cloud to be processed is first obtained, and the points whose vertical coordinates in the first point cloud are greater than the first threshold are deleted to obtain the second point cloud; then, the second point cloud is fitted using a plane fitting method to obtain a fitting plane of the road surface; finally, the first point cloud is obtained again, and the points whose distances from the first point cloud to the fitting plane are greater than the second threshold are deleted to obtain the processed first point cloud. The above process can remove most of the non-road surface points by deleting the points whose vertical coordinates in the first point cloud are greater than the first threshold, thereby reducing interference with subsequent plane fitting. In actual situations, the road surface point cloud of a single-frame point cloud can be approximated as a plane, so the second point cloud obtained after removing most of the non-road surface points can be fitted into a fitting plane of the road surface. After obtaining the fitting plane, the non-road surface points in the first point cloud can be deleted by calculating the distance between each point in the first point cloud and the fitting plane. Even for roads with undulations or ups and downs, the fitted plane of the road surface obtained after removing most of the non-road surface points is closer to the real road surface. For example, the plane fitted for an ups and downs road surface is also a plane with a slope. Therefore, filtering the non-road surface points in the original point cloud (i.e., the first point cloud without removing the non-road surface points) based on the fitted plane can improve the accuracy of removing the non-road surface points.

[0074] To facilitate understanding of the point cloud processing method provided in the embodiments of the present application, actual application scenarios are listed below.

[0075] like Figure 2 The figure is a schematic diagram of the operation flow of the point cloud processing method provided in an embodiment of the present application in an actual application scenario.

[0076] exist Figure 2 In the method, firstly, a continuous multi-frame original point cloud is obtained, and one of the frames of the original point cloud is read in turn, and the distance between the original point cloud and the previous frame of the point cloud is detected (i.e., the distance between the corresponding scene positions. If the first frame of the point cloud is currently read, the distance detection and judgment steps can be skipped); it is determined whether the distance exceeds the set threshold. If the distance does not exceed the threshold, the frame of the original point cloud is discarded, and the next frame of the point cloud in the multi-frame point cloud is returned to continue reading. If the distance exceeds the threshold, the frame of the original point cloud is converted from the radar coordinate system to the corresponding vehicle body coordinate system; then, according to the vertical coordinate (i.e., z coordinate) of the frame of the original point cloud, the non-road point (i.e., the point far away from the road) in the frame of the original point cloud is deleted to obtain the road point cloud; then, the road point cloud is used to fit the fitting plane of the road; then, the non-road point in the frame of the original point cloud is deleted using the fitting plane, thereby completing the processing of the frame of the original point cloud. After traversing and processing all the original point clouds, all the processed point clouds are converted to the same coordinate system and merged, thereby completing the splicing of the road point cloud.

[0077] In summary, the embodiment of the present application provides a point cloud processing method based on plane fitting, which can remove obstacles such as vehicles and pedestrians in point cloud data and reduce interference caused by road undulations or ups and downs. In addition, when selecting the original point cloud for processing, a selection method based on point cloud distance is provided, which can reduce the amount of calculation and storage while also obtaining a good point cloud stitching effect.

[0078] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0079] A point cloud processing method is mainly described above, and a point cloud processing device will be described below.

[0080] See also Figure 3 , an embodiment of a point cloud processing device in the embodiment of the present application includes:

[0081] The point cloud acquisition module 301 is used to acquire a first point cloud to be processed;

[0082] A first deletion module 302 is used to delete points in the first point cloud whose vertical coordinates are greater than a first threshold value to obtain a second point cloud;

[0083] A plane fitting module 303, configured to fit the second point cloud using a plane fitting method to obtain a fitting plane of the road surface;

[0084] The second deleting module 304 is used to delete points in the first point cloud whose distance from the fitting plane is greater than a second threshold, so as to obtain the processed first point cloud.

[0085] In an implementation of the embodiment of the present application, the first point cloud is any one frame of point cloud in a continuous multi-frame point cloud, and the multi-frame point cloud is acquired by a radar device on a vehicle; the point cloud processing device may further include:

[0086] A point cloud processing module, used for obtaining each frame of point cloud after processing in the multi-frame point cloud by adopting the same processing method as the first point cloud;

[0087] The point cloud stitching module is used to stitch each frame of the processed point cloud in the multiple frames of point cloud to obtain a third point cloud.

[0088] Furthermore, the point cloud acquisition module may include:

[0089] A point cloud acquisition unit, configured to acquire the first point cloud from the multi-frame point cloud;

[0090] A coordinate system conversion unit is used to convert the first point cloud from a radar coordinate system to a vehicle body coordinate system corresponding to the first point cloud.

[0091] Furthermore, the point cloud acquisition module may also include:

[0092] A distance detection unit, configured to detect a target distance between a scene position corresponding to the first point cloud and a scene position corresponding to a fourth point cloud; wherein the fourth point cloud is a previous frame point cloud adjacent to the first point cloud in the multi-frame point cloud;

[0093] An execution unit is used for executing the step of converting the first point cloud from the radar coordinate system to the vehicle body coordinate system corresponding to the first point cloud and subsequent steps if the target distance is greater than a third threshold.

[0094] Furthermore, the distance detection unit may include:

[0095] a driving distance detection subunit, configured to detect a driving distance of the vehicle from a first time point to a second time point; wherein the first time point is a time point at which the radar device collects the fourth point cloud, and the second time point is a time point at which the radar device collects the first point cloud;

[0096] The distance determination subunit is used to determine the driving distance as the target distance.

[0097] Furthermore, the point cloud acquisition module may also include:

[0098] A point cloud deleting unit is used to delete the first point cloud from the multi-frame point cloud if the target distance is less than or equal to the third threshold.

[0099] In one implementation of the embodiment of the present application, the point cloud stitching module may include:

[0100] The point cloud stitching unit is used to convert each frame of point cloud in the multi-frame point cloud into the same coordinate system after processing and perform point cloud merging processing to obtain the third point cloud; wherein the same coordinate system is the vehicle body coordinate system corresponding to any frame of point cloud in the multi-frame point cloud.

[0101] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the point cloud processing method represented by any of the above embodiments is implemented.

[0102] An embodiment of the present application also provides a computer program product. When the computer program product is run on a terminal device, the terminal device executes a point cloud processing method as represented by any of the above embodiments.

[0103] Figure 4 Schematic diagram of a terminal device provided by an embodiment of the present application. Figure 4 As shown, the terminal device 4 of this embodiment includes: a processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the processor 40. When the processor 40 executes the computer program 42, the steps in the above-mentioned embodiments of the point cloud processing method are implemented, for example Figure 1 Alternatively, when the processor 40 executes the computer program 42, the functions of each module / unit in the above-mentioned device embodiments are realized, for example, Figure 3 Functions of modules 301 to 304 are shown.

[0104] The computer program 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 42 in the terminal device 4.

[0105] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0106] The memory 41 may be an internal storage unit of the terminal device 4, such as a hard disk or memory of the terminal device 4. The memory 41 may also be an external storage device of the terminal device 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the terminal device 4. The memory 41 is used to store the computer program and other programs and data required by the terminal device. The memory 41 may also be used to temporarily store data that has been output or is to be output.

[0107] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0108] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0109] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0110] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0111] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0112] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0113] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0114] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0115] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person 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 do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A point cloud processing method, characterized in that: include: Get the first point cloud to be processed; Deleting points in the first point cloud whose vertical coordinates are greater than a first threshold value to obtain a second point cloud; Fitting the second point cloud using a plane fitting method to obtain a fitting plane of the road surface; Points in the first point cloud whose distances from the fitting plane are greater than a second threshold are deleted to obtain the processed first point cloud.

2. The method according to claim 1, characterized in that The first point cloud is any one frame of point cloud in a continuous multi-frame point cloud, and the multi-frame point cloud is acquired by a radar device on the vehicle; the method further includes: Obtaining each frame of point cloud after processing in the multi-frame point cloud by using the same processing method as the first point cloud; Each frame of point cloud processed in the multiple frames of point cloud is spliced ​​to obtain a third point cloud.

3. The method according to claim 2, characterized in that The step of obtaining a first point cloud to be processed includes: Acquire the first point cloud from the multi-frame point clouds; The first point cloud is converted from a radar coordinate system to a vehicle body coordinate system corresponding to the first point cloud.

4. The method according to claim 3, characterized in that Before converting the first point cloud from the radar coordinate system to the vehicle body coordinate system corresponding to the first point cloud, the method further includes: Detecting a target distance between a scene position corresponding to the first point cloud and a scene position corresponding to a fourth point cloud; wherein the fourth point cloud is a previous frame point cloud adjacent to the first point cloud in the multi-frame point cloud; If the target distance is greater than a third threshold, a step of converting the first point cloud from a radar coordinate system to a vehicle body coordinate system corresponding to the first point cloud and subsequent steps are performed.

5. The method according to claim 4, characterized in that The detecting a target distance between a scene position corresponding to the first point cloud and a scene position corresponding to the fourth point cloud includes: Detecting the travel distance of the vehicle from a first time point to a second time point; wherein the first time point is the time point when the radar device collects the fourth point cloud, and the second time point is the time point when the radar device collects the first point cloud; The travel distance is determined as the target distance.

6. The method according to claim 4, characterized in that After detecting the target distance between the scene position corresponding to the first point cloud and the scene position corresponding to the fourth point cloud, the method further includes: If the target distance is less than or equal to the third threshold, the first point cloud is deleted from the multi-frame point cloud.

7. The method according to any one of claims 2 to 6, characterized in that: The step of stitching each frame of the processed point cloud in the multiple frames of point cloud to obtain a third point cloud includes: Each frame of point cloud processed in the multiple frames of point cloud is converted to the same coordinate system and point cloud merging processing is performed to obtain the third point cloud; wherein the same coordinate system is the vehicle body coordinate system corresponding to any frame of point cloud in the multiple frames of point cloud.

8. A point cloud processing device, characterized in that: include: A point cloud acquisition module, used for acquiring a first point cloud to be processed; A first deletion module, used to delete points in the first point cloud whose vertical coordinates are greater than a first threshold, to obtain a second point cloud; A plane fitting module, used for fitting the second point cloud using a plane fitting method to obtain a fitting plane of the road surface; The second deletion module is used to delete points in the first point cloud whose distance from the fitting plane is greater than a second threshold, so as to obtain the processed first point cloud.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the point cloud processing method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the point cloud processing method according to any one of claims 1 to 7 is implemented.