Point cloud processing method, device, equipment and storage medium

Abstract

The present invention provides a point cloud processing method, device, equipment and storage medium. The above-mentioned point cloud processing method is applied to a laser radar, and the method includes: acquiring point cloud data; the point cloud data includes at least one point cloud; for any point cloud in the point cloud data, adjusting the first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, performing offset processing on the point cloud, and obtaining a target point cloud; wherein the first coordinate value is the distance between the laser radar represented by the point cloud and the target object in the first direction. The present invention adjusts the first coordinate value of the point cloud based on the first coefficient corresponding to the point cloud, and performs offset processing on the point cloud. The point cloud after offset processing can represent more point cloud information, thereby presenting the point cloud information to the greatest extent, thereby improving the accuracy of object detection.

Description

Technical Field

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

[0002] In the field of autonomous driving, LiDAR (LiDAR) has become an indispensable sensor for autonomous vehicles. LiDAR can be used to obtain point cloud data within a certain range of the vehicle. This point cloud data can be used as input for deep learning models to detect objects and achieve assisted driving.

[0003] Currently, point cloud data can be projected onto a two-dimensional grid map, including both a bird's-eye view and a front view, and used as input for deep learning models. However, in this approach, neither the bird's-eye view nor the front view fully captures the point cloud information, resulting in the deep learning model missing some point cloud information, which reduces object detection accuracy. Summary of the Invention

[0004] The main purpose of the present invention is to provide a point cloud processing method, device, equipment and storage medium, aiming to solve the technical problem that existing point cloud processing methods cannot present point cloud information to the greatest extent, thereby reducing the accuracy of object detection.

[0005] To achieve the above object, the present invention provides a point cloud processing method applied to laser radar, the method comprising:

[0006] Acquire point cloud data; the point cloud data includes at least one point cloud;

[0007] For any point cloud in the point cloud data, adjusting a first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, performing an offset process on the point cloud, and obtaining a target point cloud;

[0008] The first coordinate value is the distance between the laser radar represented by the point cloud and the target object in the first direction.

[0009] Optionally, the point cloud includes a second coordinate value, where the second coordinate value is the distance between the laser radar represented by the point cloud and the target object in a second direction;

[0010] Adjusting the first coordinate value of the point cloud based on the first coefficient corresponding to the point cloud and performing offset processing on the point cloud to obtain the target point cloud includes:

[0011] determining the first coefficient based on the first coordinate value and a preset second coefficient;

[0012] determining a target value as a result of multiplying the first coefficient by the second coordinate value;

[0013] Determine the sum of the target value and the first coordinate value as the first target coordinate value;

[0014] The first coordinate value of the point cloud is modified to the first target coordinate value to obtain the target point cloud.

[0015] Optionally, determining the first coefficient based on the first coordinate value and a preset second coefficient includes:

[0016] Performing a power operation on the first coordinate value to obtain a second target coordinate value;

[0017] A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

[0018] Optionally, the target point cloud includes a third coordinate value, a fourth coordinate value, and a fifth coordinate value, the third coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a first direction, the fourth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a second direction, and the fifth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a third direction;

[0019] After obtaining the target point cloud, the method includes:

[0020] Multiplying the third coordinate value, the fourth coordinate value, and the fifth coordinate value corresponding to the target point cloud by the rotation matrix respectively, and rotating the target point cloud;

[0021] The rotation matrix is ​​determined based on a preset rotation angle.

[0022] In addition, to achieve the above-mentioned purpose, the present invention further provides a point cloud processing device, which is applied to a laser radar and includes:

[0023] An acquisition module, configured to acquire point cloud data; the point cloud data includes at least one point cloud;

[0024] a first processing module, configured to adjust, for any point cloud in the point cloud data, a first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, and perform an offset process on the point cloud to obtain a target point cloud;

[0025] The first coordinate value is the distance between the laser radar represented by the point cloud and the target object in the first direction.

[0026] Optionally, the point cloud includes a second coordinate value, where the second coordinate value is the distance between the laser radar represented by the point cloud and the target object in a second direction;

[0027] The first processing module is specifically configured to:

[0028] determining the first coefficient based on the first coordinate value and a preset second coefficient;

[0029] determining a target value as a result of multiplying the first coefficient by the second coordinate value;

[0030] Determine the sum of the target value and the first coordinate value as the first target coordinate value;

[0031] The first coordinate value of the point cloud is modified to the first target coordinate value to obtain the target point cloud.

[0032] Optionally, the first processing module is further configured to:

[0033] Performing a power operation on the first coordinate value to obtain a second target coordinate value;

[0034] A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

[0035] Optionally, the target point cloud includes a third coordinate value, a fourth coordinate value, and a fifth coordinate value, the third coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a first direction, the fourth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a second direction, and the fifth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a third direction;

[0036] The device further comprises:

[0037] A second processing module is configured to perform multiplication operations on the third coordinate value, the fourth coordinate value, and the fifth coordinate value corresponding to the target point cloud with the rotation matrix respectively, so as to perform rotation processing on the target point cloud;

[0038] The rotation matrix is ​​determined based on a preset rotation angle.

[0039] In addition, to achieve the above-mentioned purpose, the present invention also provides an electronic device, which includes a memory, a processor, and a program or instruction stored in the memory and runnable on the processor, and the processor implements the steps of the point cloud processing method described above when executing the program or instruction.

[0040] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, on which a data processing program is stored. When the data processing program is executed by a processor, the steps of the point cloud processing method described above are implemented.

[0041] The present invention provides a point cloud processing method, apparatus, device, and storage medium. The above-mentioned point cloud processing method includes: acquiring point cloud data; the point cloud data includes at least one point cloud; for any point cloud in the point cloud data, adjusting the first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, performing offset processing on the point cloud, and obtaining a target point cloud; wherein the first coordinate value is the distance between the laser radar represented by the point cloud and the target object in a first direction. In the present invention, the first coordinate value of the point cloud is adjusted based on the first coefficient corresponding to the point cloud, and the point cloud is offset. The offset-processed point cloud can represent more point cloud information, thereby presenting the point cloud information to the greatest extent, thereby improving the accuracy of object detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 Schematic diagram of the hardware operating environment of the embodiment of the present invention;

[0043] Figure 2 It is a schematic diagram of projecting point cloud data in the prior art;

[0044] Figure 3 1 is a flow chart of a point cloud processing method provided by an embodiment of the present invention;

[0045] Figure 4 This is one of the schematic diagrams after projecting point cloud data using the point cloud processing method provided by an embodiment of the present invention;

[0046] Figure 5 This is a second schematic diagram after projecting point cloud data using the point cloud processing method provided by an embodiment of the present invention;

[0047] Figure 6 1 is a schematic structural diagram of a point cloud processing device provided by an embodiment of the present invention;

[0048] Figure 7 It is a structural diagram of an electronic device provided by an embodiment of the present invention.

[0049] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0051] like Figure 1 As shown, Figure 1 It is a schematic diagram of the terminal structure of the hardware operating environment involved in the embodiment of the present invention.

[0052] The terminal of the present invention is a movable device, and the terminal can also be other terminal devices with storage functions.

[0053] like Figure 1 As shown, the terminal may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and optionally the user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may optionally also be a storage device independent of the aforementioned processor 1001.

[0054] Optionally, the terminal may further include a camera, a Wi-Fi module, etc., which will not be described in detail here.

[0055] Those skilled in the art will understand that Figure 1 The terminal structure shown in the figure does not constitute a limitation to the terminal, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0056] exist Figure 1 In the terminal shown, the network interface 1004 is mainly used to connect to the backend server and communicate data with the backend server; the user interface 1003 mainly includes an input unit such as a keyboard, including a wireless keyboard and a wired keyboard, which is used to connect to the client and communicate data with the client; and the processor 1001 can be used to call the data processing program stored in the memory 1005 and perform the following operations:

[0057] Get point cloud data;

[0058] For any point cloud in the point cloud data, a first coordinate value of the point cloud is adjusted based on a first coefficient corresponding to the point cloud, and an offset process is performed on the point cloud to obtain a target point cloud.

[0059] Furthermore, the processor 1001 may call the data processing program stored in the memory 1005 and perform the following operations:

[0060] determining the first coefficient based on the first coordinate value and a preset second coefficient;

[0061] determining a target value as a result of multiplying the first coefficient by the second coordinate value;

[0062] Determine the sum of the target value and the first coordinate value as the first target coordinate value;

[0063] The first coordinate value of the point cloud is modified to the first target coordinate value to obtain the target point cloud.

[0064] Furthermore, the processor 1001 may call the data processing program stored in the memory 1005 and perform the following operations:

[0065] Performing a power operation on the first coordinate value to obtain a second target coordinate value;

[0066] A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

[0067] Furthermore, the processor 1001 may call the data processing program stored in the memory 1005 and perform the following operations:

[0068] The third coordinate value, the fourth coordinate value, and the fifth coordinate value corresponding to the target point cloud are multiplied by the rotation matrix respectively, and the target point cloud is rotated.

[0069] The specific embodiments of this terminal are basically the same as the embodiments of the point cloud processing method described below, and will not be described in detail here.

[0070] The point cloud processing method provided in this embodiment is applied to a laser radar. Optionally, the laser radar can be a vehicle-mounted laser radar, which is used to collect point cloud data and is typically mounted on the roof or front of the vehicle. The collected point cloud data is used as input for a deep learning model to implement functions such as point cloud object detection, simultaneous localization and mapping, multi-sensor calibration, active area detection, and trajectory prediction.

[0071] Currently, the following three methods are commonly used to process point cloud data and use the processed point cloud data as input for deep learning models.

[0072] The first method is pure point cloud input:

[0073] Pure point cloud input retains the complete three-dimensional spatial features of the point cloud without information loss, which is conducive to object detection. However, due to the large number of points in the point cloud, point-based feature extraction is very time-consuming, so the use scenarios of the above point cloud processing method are limited.

[0074] The second method is to convert the point cloud into voxel input:

[0075] Converting point clouds into voxel inputs can retain some of the three-dimensional spatial features of the point clouds, but the above method is difficult to deploy and it is difficult to meet real-time requirements.

[0076] The third method is to project the point cloud into a two-dimensional grid image and use the grid image as input:

[0077] In the above method, the two-dimensional grid map includes a bird's-eye view and a front view, but the two-dimensional grid map cannot present the point cloud information to the greatest extent, resulting in the input of the deep learning model missing some point cloud information, which reduces the accuracy of object detection.

[0078] For easier understanding, see Figure 2 , Figure 2 This is a schematic diagram of projecting point cloud data in the prior art. Figure 2 The bird's-eye view corresponding to the point cloud data can be obtained, but the height information of the point cloud will be lost.

[0079] In other words, the existing point cloud processing method cannot present point cloud information to the greatest extent, thereby reducing the technical problem of object detection accuracy.

[0080] In order to solve the above technical problems, the present invention provides a point cloud processing method. Figure 3 , Figure 3 FIG1 is a flow chart of a point cloud processing method provided by an embodiment of the present invention. The point cloud processing method provided by this embodiment is applied to laser radar.

[0081] The point cloud processing method provided by the embodiment of the present invention includes the following steps:

[0082] S110, acquiring point cloud data.

[0083] In this step, the point cloud data can be obtained by data transmission, downloading the point cloud data from the Internet, or other methods. The specific method of obtaining the point cloud data is not limited here.

[0084] The above-mentioned point cloud data is data collected by a laser radar, wherein the point cloud data includes at least one point cloud.

[0085] S120 , for any point cloud in the point cloud data, adjust the first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, perform offset processing on the point cloud, and obtain a target point cloud.

[0086] The first coordinate value is the distance in the first direction between the laser radar represented by the point cloud and the target object. In some optional embodiments, the target object may be an object or a human body detected by the laser radar. For ease of understanding, a three-dimensional rectangular coordinate system can be established with the laser radar as the center, where the Y-axis of the coordinate system represents the forward direction of the vehicle equipped with the laser radar, i.e., the first direction.

[0087] The first coefficient is related to the first coordinate value. For a specific technical solution on how to determine the first coefficient, please refer to the subsequent embodiments.

[0088] In this step, after obtaining the first coefficient of any point cloud, the first coefficient is used to adjust the first coordinate value of the point cloud, that is, to adjust the distance between the laser radar represented by the point cloud and the target object in the first direction, to achieve offset processing of the point cloud and obtain the target point cloud. The target point cloud can better represent the height information compared to the point cloud before offset processing.

[0089] For easier understanding, see Figure 4 , Figure 4 This is one of the schematic diagrams after projecting point cloud data using the point cloud processing method provided by the embodiment of the present invention. Figure 4 FIG. 1 is a schematic diagram of a target point cloud after performing offset processing on all point clouds in the point cloud data.

[0090] The point cloud processing method provided by the present invention includes: acquiring point cloud data; the point cloud data includes at least one point cloud; for any point cloud in the point cloud data, adjusting a first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, performing offset processing on the point cloud, and obtaining a target point cloud; wherein the first coordinate value is the distance in a first direction between a laser radar represented by the point cloud and a target object. In the present invention, the first coordinate value of the point cloud is adjusted based on the first coefficient corresponding to the point cloud, and the point cloud is offset. The offset point cloud can represent more point cloud information, thereby presenting the point cloud information to the greatest extent possible, thereby improving the accuracy of object detection.

[0091] Optionally, the point cloud includes a second coordinate value, where the second coordinate value is the distance between the laser radar represented by the point cloud and the target object in a second direction;

[0092] Adjusting the first coordinate value of the point cloud based on the first coefficient corresponding to the point cloud and performing offset processing on the point cloud to obtain the target point cloud includes:

[0093] determining the first coefficient based on the first coordinate value and a preset second coefficient;

[0094] determining a target value as a result of multiplying the first coefficient by the second coordinate value;

[0095] Determine the sum of the target value and the first coordinate value as the first target coordinate value;

[0096] The first coordinate value of the point cloud is modified to the first target coordinate value to obtain the target point cloud.

[0097] As described above, a three-dimensional rectangular coordinate system can be established with the lidar as the center, where the Z-axis of the coordinate system represents the height direction of the vehicle on which the lidar is mounted, i.e., the second direction. The distance between the lidar represented by the point cloud and the target object in the second direction is referred to as the second coordinate value. In this embodiment, the second coordinate value can be directly obtained.

[0098] The above-mentioned second coefficient can be the radar vertical angular resolution. It should be understood that the above-mentioned radar vertical angular resolution is a radar performance parameter determined during the design of the laser radar and is a preset value.

[0099] In this embodiment, the first coefficient can be determined based on the first coordinate value and the second coefficient. For a specific technical solution on how to determine the first coefficient, please refer to the subsequent embodiments.

[0100] After obtaining the first coefficient, the first coefficient is multiplied by the second coordinate value to obtain a target value, and the target value is added to the first coordinate value to obtain a first target coordinate value. The first target coordinate value is the distance between the laser radar and the target object in the first direction represented by the offset point cloud.

[0101] For easier understanding, see the following formula:

[0102] y2=y1+αz1

[0103] Among them, y2 is the first target coordinate value, y1 is the first coordinate value, α is the first coefficient, and z1 is the second coordinate value.

[0104] Optionally, determining the first coefficient based on the first coordinate value and a preset second coefficient includes:

[0105] Performing a power operation on the first coordinate value to obtain a second target coordinate value;

[0106] A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

[0107] The first coefficient can be calculated by the following formula:

[0108]

[0109] Among them, α is the first coefficient, y1 is the first coordinate value, and β is the second coefficient.

[0110] In this embodiment, the third power of the first coordinate value is used as the second target coordinate value, and the second target coordinate value is multiplied by the square of the second coefficient to obtain the first coefficient.

[0111] Then, based on the above embodiment, the first target coordinate value can be calculated using the following formula:

[0112]

[0113] Among them, y2 is the first target coordinate value, y1 is the first coordinate value, β is the second coefficient, and z1 is the second coordinate value.

[0114] Optionally, the target point cloud includes a third coordinate value, a fourth coordinate value, and a fifth coordinate value, the third coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a first direction, the fourth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a second direction, and the fifth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a third direction;

[0115] After obtaining the target point cloud, the method includes:

[0116] The third coordinate value, the fourth coordinate value, and the fifth coordinate value corresponding to the target point cloud are multiplied by the rotation matrix respectively, and the target point cloud is rotated.

[0117] It should be understood that after the point cloud is offset to obtain the target point cloud, the feature enhancement effect brought about by the offset of the point cloud is limited. On this basis, we can rotate the target point cloud so that the rotated point cloud includes more point cloud information.

[0118] In this embodiment, a rectangular three-dimensional coordinate system can be established with the target point cloud as the center, wherein the Y-axis of the coordinate system represents the forward direction of the vehicle equipped with the laser radar, that is, the first direction, and the distance between the laser radar represented by the target point cloud and the target object in the first direction is called the third coordinate value; the Z-axis of the coordinate system represents the height direction of the vehicle equipped with the laser radar, that is, the second direction, and the distance between the laser radar represented by the target point cloud and the target object in the second direction is called the fourth coordinate value; the X-axis of the coordinate system represents the horizontal direction of the vehicle equipped with the laser radar, that is, the third direction, and the distance between the laser radar represented by the target point cloud and the target object in the third direction is called the fifth coordinate value.

[0119] In this embodiment, a rotation matrix is ​​pre-set, and the rotation matrix is ​​expressed as follows:

[0120]

[0121] Where θ represents the rotation angle, which is a preset value and is related to the installation height and scanning mode of the LiDAR. For example, when the LiDAR is 40cm above the ground, the rotation angle can be set between 15 degrees and 45 degrees.

[0122] In this embodiment, the third coordinate value, the fourth coordinate value, and the fifth coordinate value are multiplied by the rotation matrix respectively to obtain the coordinates of the target point cloud after the rotation processing, thereby realizing the rotation processing of the target point cloud.

[0123] For easier understanding, see the following formula:

[0124] P1(x1,y1,z1)=R x (θ)*P(x,y,z)

[0125] Among them, P1(x1,y1,z1) represents the coordinates of the target point cloud after rotation processing, R x (θ) represents the rotation matrix, and P(x, y, z) represents the coordinates of the target point cloud.

[0126] See also Figure 5 , Figure 5 This is a second schematic diagram of projecting point cloud data using the point cloud processing method provided by an embodiment of the present invention. Figure 5 The diagram shows a schematic diagram after the target point cloud is rotated.

[0127] In addition, the present invention also provides a point cloud processing device, see Figure 6 , Figure 6 Schematic diagram of the structure of a point cloud processing device provided by an embodiment of the present invention.

[0128] like Figure 6 As shown, the point cloud processing device 200 includes:

[0129] An acquisition module 210 is used to acquire point cloud data;

[0130] The first processing module 220 is configured to adjust the first coordinate value of any point cloud in the point cloud data based on a first coefficient corresponding to the point cloud, perform offset processing on the point cloud, and obtain a target point cloud.

[0131] Optionally, the point cloud includes a second coordinate value, where the second coordinate value is the distance between the laser radar represented by the point cloud and the target object in a second direction;

[0132] The first processing module 220 is specifically configured to:

[0133] determining the first coefficient based on the first coordinate value and a preset second coefficient;

[0134] determining a target value as a result of multiplying the first coefficient by the second coordinate value;

[0135] Determine the sum of the target value and the first coordinate value as the first target coordinate value;

[0136] The first coordinate value of the point cloud is modified to the first target coordinate value to obtain the target point cloud.

[0137] Optionally, the first processing module 220 is further configured to:

[0138] Performing a power operation on the first coordinate value to obtain a second target coordinate value;

[0139] A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

[0140] Optionally, the target point cloud includes a third coordinate value, a fourth coordinate value, and a fifth coordinate value, the third coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a first direction, the fourth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a second direction, and the fifth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in a third direction;

[0141] The point cloud processing device 200 further includes:

[0142] The second processing module is used to perform multiplication operations on the third coordinate value, the fourth coordinate value and the fifth coordinate value corresponding to the target point cloud with the rotation matrix respectively, so as to perform rotation processing on the target point cloud.

[0143] The point cloud processing device 200 can realize the embodiment of the present invention Figure 2 In the various processes of the method embodiment, in this embodiment, the first coordinate value of the point cloud is adjusted based on the first coefficient corresponding to the point cloud, and the point cloud is offset. The point cloud after offset processing can represent more point cloud information, thereby presenting the point cloud information to the greatest extent, thereby improving the accuracy of object detection.

[0144] In addition, an embodiment of the present invention further proposes a computer-readable storage medium, on which a data processing program is stored. When the data processing program is executed by a processor, each step in the above-mentioned point cloud processing method is implemented, which will not be repeated here.

[0145] The specific embodiments of the computer-readable storage medium of the present invention are basically the same as the embodiments of the above-mentioned point cloud processing method, and are not described in detail here.

[0146] The embodiment of the present invention further provides an electronic device, such as Figure 7 As shown, it includes a processor 310 , a communication interface 320 , a memory 330 and a communication bus 340 , wherein the processor 310 , the communication interface 320 , and the memory 330 communicate with each other via the communication bus 340 .

[0147] The memory 330 is used to store computer programs;

[0148] The processor 310 is configured to execute a program stored in the memory 330 , wherein the computer program is configured to obtain point cloud data when executed by the processor 310 ;

[0149] For any point cloud in the point cloud data, a first coordinate value of the point cloud is adjusted based on a first coefficient corresponding to the point cloud, and an offset process is performed on the point cloud to obtain a target point cloud.

[0150] When the computer program is executed by the processor 310, it is configured to determine the first coefficient based on the first coordinate value and a preset second coefficient;

[0151] determining a target value as a result of multiplying the first coefficient by the second coordinate value;

[0152] Determine the sum of the target value and the first coordinate value as the first target coordinate value;

[0153] The first coordinate value of the point cloud is modified to the first target coordinate value to obtain the target point cloud.

[0154] When the processor 310 executes the computer program, the computer program is configured to perform a power operation on the first coordinate value to obtain a second target coordinate value;

[0155] A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

[0156] When the computer program is executed by the processor 310 , it is used to perform multiplication operations on the third coordinate value, the fourth coordinate value, and the fifth coordinate value corresponding to the target point cloud with the rotation matrix respectively, so as to rotate the target point cloud.

[0157] The communication bus mentioned in the electronic device mentioned above may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0158] The communication interface is used for communication between the above terminal and other devices.

[0159] The memory may include random access memory (RAM) or non-volatile memory (non-volatile memory), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0160] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0161] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0162] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0163] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0164] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A point cloud processing method, characterized in that: Applied to laser radar, the method includes: Acquire point cloud data; the point cloud data includes at least one point cloud; For any point cloud in the point cloud data, adjusting a first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, performing an offset process on the point cloud, and obtaining a target point cloud; Wherein, the first coordinate value is the distance between the laser radar and the target object in the first direction represented by the point cloud; The point cloud includes a second coordinate value, where the second coordinate value is a distance between the laser radar and the target object represented by the point cloud in a second direction; Adjusting the first coordinate value of the point cloud based on the first coefficient corresponding to the point cloud and performing offset processing on the point cloud to obtain the target point cloud includes: determining the first coefficient based on the first coordinate value and a preset second coefficient; determining a target value as a result of multiplying the first coefficient by the second coordinate value; Determine the sum of the target value and the first coordinate value as the first target coordinate value; Modifying the first coordinate value of the point cloud to the first target coordinate value to obtain the target point cloud; The determining the first coefficient based on the first coordinate value and a preset second coefficient includes: Performing a power operation on the first coordinate value to obtain a second target coordinate value; A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

2. The method according to claim 1, characterized in that The target point cloud includes a third coordinate value, a fourth coordinate value, and a fifth coordinate value, the third coordinate value is the distance between the laser radar represented by the target point cloud and the target object in the first direction, the fourth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in the second direction, and the fifth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in the third direction; After obtaining the target point cloud, the method includes: Multiplying the third coordinate value, the fourth coordinate value, and the fifth coordinate value corresponding to the target point cloud by the rotation matrix respectively, and rotating the target point cloud; The rotation matrix is ​​determined based on a preset rotation angle.

3. A point cloud processing device, characterized in that: Applied to laser radar, the device includes: An acquisition module, configured to acquire point cloud data; the point cloud data includes at least one point cloud; a first processing module, configured to adjust, for any point cloud in the point cloud data, a first coordinate value of the point cloud based on a first coefficient corresponding to the point cloud, and perform an offset process on the point cloud to obtain a target point cloud; The first coordinate value is the distance between the laser radar represented by the point cloud and the target object in the first direction, and the point cloud includes a second coordinate value, and the second coordinate value is the distance between the laser radar represented by the point cloud and the target object in the second direction; The first processing module is specifically configured to: determining the first coefficient based on the first coordinate value and a preset second coefficient; determining a target value as a result of multiplying the first coefficient by the second coordinate value; Determine the sum of the target value and the first coordinate value as the first target coordinate value; Modifying the first coordinate value of the point cloud to the first target coordinate value to obtain the target point cloud; The first processing module is further specifically configured to: Performing a power operation on the first coordinate value to obtain a second target coordinate value; A multiplication result of the second target coordinate value and the square of the second coefficient is determined as the first coefficient.

4. The device according to claim 3, characterized in that The target point cloud includes a third coordinate value, a fourth coordinate value, and a fifth coordinate value, the third coordinate value is the distance between the laser radar represented by the target point cloud and the target object in the first direction, the fourth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in the second direction, and the fifth coordinate value is the distance between the laser radar represented by the target point cloud and the target object in the third direction; The device further comprises: A second processing module is configured to perform multiplication operations on the third coordinate value, the fourth coordinate value, and the fifth coordinate value corresponding to the target point cloud with the rotation matrix respectively, so as to perform rotation processing on the target point cloud; The rotation matrix is ​​determined based on a preset rotation angle.

5. An electronic device comprising a memory, a processor, and a program or instruction stored in the memory and executable on the processor, wherein: When the processor executes the program or pointer, the steps of the point cloud processing method according to claim 1 or 2 are implemented.

6. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by a processor, the steps of the point cloud processing method according to claim 1 or 2 are implemented.

Images