Noise Data Identification Method, Device and Storage Medium

Abstract

The present invention discloses a method, device and storage medium for identifying noise data. The method includes: obtaining point cloud data of a lidar, and segmenting the point cloud data through a grid; determining a first grid and a second grid successively passed by the emission direction of the lidar; calculating the height range of the point cloud data in the first grid; and when there are target point clouds in the second grid whose heights are related to the height range, determining that the target point clouds are noise data. The present invention aims to improve the recognition rate of crosstalk noise data.

Description

Technical Field

[0001] The present invention relates to the technical field of radar, and in particular to a method, device and storage medium for identifying noise data. Background Art

[0002] As a technology that can obtain accurate three-dimensional space, lidar has the advantages of being less affected by natural conditions such as weather and light, and is widely used in the field of driverless. However, in actual life applications, radar signals will affect each other and generate noise. The traditional radar denoising method is to add codes to the emitted laser signals, and then decode the received radar data to solve the crosstalk noise. However, this cannot completely solve the problem of crosstalk noise. For example, when a vehicle A is driving, it will encounter another vehicle B equipped with lidar. The signals generated by the lidar of vehicle B will affect the signal reception of vehicle A. The radar data received by the lidar of vehicle A will have crosstalk noise. However, the coding information of the laser signals emitted by vehicle B may be the same as that of vehicle A, and the crosstalk noise of vehicle B cannot be identified. Therefore, the traditional noise data identification method has a low recognition rate for crosstalk noise data.

[0003] The above content is only used to assist in understanding the technical solution of the present invention, and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method, device and storage medium for identifying noise data, aiming to improve the recognition rate of crosstalk noise data.

[0005] To achieve the above object, the present invention provides a method for identifying noise data, and the method for identifying noise data includes the following steps:

[0006] Obtain the point cloud data of the lidar, and segment the point cloud data through a grid;

[0007] Determine the first grid and the second grid that the lidar emission direction passes through in sequence;

[0008] Calculate the height range of the point cloud data in the first grid;

[0009] When there is target point cloud in the second grid whose height is related to the height range, determine that the target point cloud is noise data.

[0010] Optionally, the determining the first grid and the second grid that the lidar emission direction passes through in sequence includes:

[0011] Traverse the grid along the lidar emission direction. If there is a target grid where the number of laser points is greater than or equal to a preset number, determine the target grid as the first grid;

[0012] Starting from the first grid and with the emission direction of the lidar as the extension direction, the second grid is determined.

[0013] Optionally, the step of starting from the first grid and with the emission direction of the lidar as the extension direction to determine the second grid includes:

[0014] Starting from the first grid and with the emission direction of the lidar as the extension direction, a preset number of grids passed through by the extension are respectively used as the second grid.

[0015] Optionally, the step of determining the target point cloud as noise data when there is a target point cloud with a height related to the height range in the second grid includes:

[0016] Traverse a preset number of the second grids. If there is a target point cloud with a height within the height range, determine the target point cloud as noise data.

[0017] Optionally, the method further includes:

[0018] Traverse a preset number of the second grids. If the ratio of the number of laser points in the first grid to the number of laser points in the second grid is greater than or equal to a preset ratio, then determine the point cloud data in the second grid as noise data.

[0019] Optionally, the step of determining the target point cloud as noise data when there is a target point cloud with a height related to the height range in the second grid includes:

[0020] Calculate the height difference between the first grid and the second grid;

[0021] If the sum of the height of the laser points in the second grid and the height difference is within the height range, then the laser points are the target point cloud, and determine the target point cloud as noise data.

[0022] Optionally, the step of obtaining the point cloud data of the lidar and segmenting the point cloud data by grids includes:

[0023] Obtain the point cloud data of the lidar, and determine the boundary information and grid resolution of the point cloud data;

[0024] Determine grid parameters according to the boundary information and the grid resolution, where the grid parameters include grid shape and grid size;

[0025] Segment the point cloud data according to the grid parameters.

[0026] Optionally, after determining the target point cloud as noise data when there is a target point cloud with a height related to the height range in the second grid, it further includes:

[0027] Remove the target point cloud.

[0028] In addition, to achieve the above object, the present invention further provides a noise data recognition device, which includes a memory, a processor, and a noise data recognition program stored on the memory and executable on the processor. When the noise data recognition program is executed by the processor, the steps of the above-described noise data recognition method are implemented.

[0029] In addition, to achieve the above object, the present invention further provides a noise data recognition device, which includes:

[0030] An acquisition module, configured to acquire the point cloud data of the lidar and segment the point cloud data through a grid;

[0031] A determination module, configured to determine a first grid and a second grid that are sequentially passed by the lidar emission direction;

[0032] A calculation module, configured to calculate the height range of the point cloud data in the first grid;

[0033] A determination module, configured to determine that the target point cloud is noise data when there is a target point cloud in the second grid whose height is related to the height range.

[0034] In addition, to achieve the above object, the present invention further provides a computer-readable storage medium, on which a noise data recognition program is stored. When the noise data recognition program is executed by a processor, the steps of the above-described noise data recognition method are implemented.

[0035] A noise data recognition method, device, and storage medium provided by an embodiment of the present invention first acquire the point cloud data of the lidar and segment the point cloud data through a grid; determine a first grid and a second grid that are sequentially passed by the lidar emission direction; calculate the height range of the point cloud data in the first grid; when there is a target point cloud in the second grid whose height is related to the height range, determine that the target point cloud is noise data. The acquired lidar point cloud data is segmented through a grid, and the second grid located after the first grid is determined according to the laser emission direction. Since the laser cannot detect the data inside or behind the object, when there is a target point cloud in the second grid whose height is related to the height range, it is determined that the target point cloud is noise data, thereby improving the recognition rate of crosstalk noise data. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of the terminal structure of the hardware operating environment involved in the solution of the embodiment of the present invention;

[0037] Figure 2Schematic flowchart of an embodiment of the noise data recognition method of the present invention;

[0038] Figure 3 Schematic diagram of an application scenario of an embodiment related to the embodiment of the present invention.

[0039] The realization, functional characteristics and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0040] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0041] In the related art, the laser signals from other systems cannot be completely identified, and the laser signals from other systems will cause crosstalk noise in the point cloud data of this system. The traditional noise data recognition method cannot fundamentally identify the crosstalk noise, resulting in a low recognition rate of the crosstalk noise data.

[0042] In order to improve the recognition rate of the crosstalk noise data, the embodiments of the present invention propose a noise data recognition method, device and storage medium. The main steps of the method include:

[0043] Obtain the point cloud data of the lidar and segment the point cloud data through grids;

[0044] Determine the first grid and the second grid that the lidar emission direction passes through in sequence;

[0045] Calculate the height range of the point cloud data in the first grid;

[0046] When there is target point cloud in the second grid whose height is related to the height range, determine that the target point cloud is noise data.

[0047] In this way, by segmenting the lidar point cloud data through grids and determining the first grid and the second grid that the laser passes through in sequence according to the laser emission direction, since the laser cannot detect the data inside or behind the object, there should be no point cloud data in the second grid that is related to the height range of the point cloud data in the first grid. When there is target point cloud in the second grid whose height is related to the height range, determine that the target point cloud is noise data. Therefore, based on the solution given in the above embodiment, directly performing noise data recognition on the point cloud data can improve the recognition rate of the crosstalk noise data.

[0048] The technical solution content of the present invention will be described in detail below with reference to the accompanying drawings.

[0049] As Figure 1 shown, Figure 1 is a schematic diagram of the terminal structure of the hardware operating environment related to the embodiment of the present invention.

[0050] The terminal in the embodiment of the present invention may be a noise data recognition device.

[0051] As Figure 1 shown, the terminal may include: a processor 1001, such as a CPU, a memory 1003, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The memory 1003 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. Optionally, the memory 1003 may also be a storage device independent of the aforementioned processor 1001.

[0052] Those skilled in the art can understand that Figure 1 the terminal structure shown in

[0053] does not constitute a limitation on the terminal, and may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements. Figure 1 As shown, the memory 1003, as a computer storage medium, may include an operating system and a noise data recognition program.

[0054] In Figure 1 the terminal shown, the processor 1001 may be used to call the noise data recognition program stored in the memory 1003 and perform the following operations:

[0055] Obtain the point cloud data of the lidar, and segment the point cloud data through a grid;

[0056] Determine a first grid and a second grid that the lidar emission direction passes through in sequence;

[0057] Calculate the height range of the point cloud data in the first grid;

[0058] When there is target point cloud in the second grid whose height is related to the height range, determine that the target point cloud is noise data.

[0059] Further, the processor 1001 may call the noise data recognition program stored in the memory 1003 and further perform the following operations:

[0060] Traverse the grid along the lidar emission direction. If there is a target grid where the number of laser points is greater than or equal to a preset number, determine the target grid as the first grid;

[0061] Starting from the first grid and with the lidar emission direction as the extension direction, determine the second grid.

[0062] Further, the processor 1001 may call the noise data recognition program stored in the memory 1003 and further perform the following operations:

[0063] Taking the first grid as the starting point and the laser radar emission direction as the extension direction, a preset number of grids passed through by the extension are respectively used as the second grids.

[0064] Further, the processor 1001 may call the noise data recognition program stored in the memory 1003 and further perform the following operations:

[0065] Traverse a preset number of the second grids. If there are target point clouds with heights within the height range, determine that the target point clouds are noise data.

[0066] Further, the processor 1001 may call the noise data recognition program stored in the memory 1003 and further perform the following operations:

[0067] Traverse a preset number of the second grids. If the ratio of the number of laser points in the first grid to the number of laser points in the second grid is greater than or equal to a preset ratio, then determine that the point cloud data in the second grid is noise data.

[0068] Further, the processor 1001 may call the noise data recognition program stored in the memory 1003 and further perform the following operations:

[0069] Calculate the height difference between the first grid and the second grid;

[0070] If the sum of the height of the laser points in the second grid and the height difference is within the height range, then the laser points are target point clouds, and determine that the target point clouds are noise data.

[0071] Further, the processor 1001 may call the noise data recognition program stored in the memory 1003 and further perform the following operations:

[0072] Obtain the point cloud data of the lidar, and determine the boundary information and grid resolution of the point cloud data;

[0073] Determine grid parameters according to the boundary information and the grid resolution, where the grid parameters include grid shape and grid size;

[0074] Segment the point cloud data according to the grid parameters.

[0075] Exemplarily, referring to Figure 2 , in an embodiment of the noise data recognition method of the present invention, the noise data recognition method includes the following steps:

[0076] Step S10: Obtain the point cloud data of the lidar, and segment the point cloud data through a grid.

[0077] In this embodiment, first obtain the lidar point cloud data that needs to be processed for noise identification. The lidar point cloud data is composed of a number of laser point data. The lidar emits a laser beam and receives the laser beam reflected from the surface of an object, thereby forming a laser point data. The point cloud data includes the angular information and distance information between the lidar and surrounding objects. In some embodiments, the point cloud data is segmented by a cube grid.

[0078] In this embodiment, the robot is equipped with a multi-line lidar. During the movement of the robot, the multi-line lidar can scan the external environment to obtain the point cloud data of the external environment, and the point cloud data is distributed in three dimensions in space.

[0079] Optionally, obtain the point cloud data of the lidar, determine the boundary information and grid resolution of the point cloud data; determine the grid parameters according to the boundary information and the grid resolution, where the grid parameters include the grid shape and grid size; segment the point cloud data according to the grid parameters.

[0080] According to the obtained point cloud data, detect the density of the point cloud data and determine the boundary information of the point cloud data, that is, most of the laser points are distributed in the space within the point cloud boundary, and there are no laser points or scattered and extremely few laser points in the space outside the point cloud boundary. In this embodiment, each grid is a cube. Of course, according to the point cloud distribution, multiple grids can be merged into a new grid to form grids with different resolutions. The shape of the grid can be a cube or a cuboid. According to the operation requirements, determine the grid resolution. For example, the resolution of the grid is 3 cm × 3 cm × 3 cm. The grid resolution cannot be too large to avoid too many laser points in one grid. The grid resolution cannot be too small to avoid too few laser points in one grid. In space, the point cloud data is segmented into several three-dimensional grids by intersecting straight lines to facilitate traversing the laser points in each grid.

[0081] Step S20: Determine the first grid and the second grid that the lidar emission direction passes through in sequence.

[0082] In this embodiment, the point cloud data is formed by the lidar emitting a laser towards the tracking point. The laser is emitted in a straight line and will pass through multiple grids in sequence. First, in one direction of the laser emission, determine whether there is a first grid that meets the preset conditions. Secondly, if the first grid exists, determine the second grid in the back direction of the first grid. After segmenting the point cloud data into multiple grids, not all grids can be used as the first grid.

[0083] Optionally, traverse the grid along the lidar emission direction. If there is a target grid where the number of laser points is greater than or equal to a preset number, determine the target grid as the first grid; starting from the first grid and with the lidar emission direction as the extension direction, determine the second grid.

[0084] Specifically, along the laser emission direction, one grid calculation unit by one, when the number of laser points in the grid is greater than or equal to the preset number, determine the first grid as the first grid. When the number of laser points in the grid reaches the preset number, it is determined as the first grid, and the laser points in the first grid are determined to be the laser points reflected by the object surface. The preset number is determined according to the actual operating scenario. If there are objects with high light transmittance in the environment, such as glass, the preset number can be smaller. Based on the first grid, with the laser emission direction as the extension direction, extending to the next grid is the second grid, and the first grid and the second grid are adjacent. In some other embodiments, the first grid and the second grid are not adjacent, and there is one or more grids between the first grid and the second grid, such as one or two.

[0085] Further, starting from the first grid and with the laser emission direction as the extension direction, extending through one or more grids, each of the one or more grids is respectively determined as the second grid.

[0086] It should be understood that the number of second grids can be one or more, but all the grids in the second grids are the grids determined by the first grid along the laser emission direction.

[0087] For better understanding, please refer to Figure 3 , the lidar point cloud data is divided into 1, 2, 3, 4, 5, 6, 7 small grids by the grid, corresponding to the actual scenario. The radar device 10 emits laser signals in three emission directions a, b, c. If grid 2 is the target grid where the number of point cloud data is greater than the preset number, determine grid 2 as the first grid. Starting from grid 2 and with the laser emission direction b as the extension direction, determine the corresponding second grid according to the laser emission direction b. If it extends through one grid, grid 5 is the second grid. If it extends through multiple grids, grid 5 and grid 8 are the second grids.

[0088] Step S20: Calculate the height range of the point cloud data in the first grid;

[0089] In this embodiment, the first grid is the grid closer to the radar emission module relative to the second grid, and there are more than a preset number of laser points in the first grid. The point cloud data is the angular information and distance information between the lidar and the laser points. According to the angular information and distance information, the height information of the laser points can be calculated. The height range of the point cloud data in the first grid is from the height of the lowest laser point in the first grid to the height of the highest laser point in the first grid, that is, the range between the lowest laser point and the highest laser point is the height range of the point cloud data in the grid.

[0090] Optionally, before the step of determining the height range of the point cloud data in the first grid, the number of laser points in the first grid should also be confirmed; when the number of laser points is less than the preset number, it is determined that the data in the first grid is not the noise data. It can be understood that if the number of laser points in the grid is not high, it is generally because the laser points are not dense enough. In the case where the laser points are not very dense, the actual scene corresponding to the grid may not have an object that can block the laser. There is point cloud in the second grid after the laser emission direction whose height is related to the height range of the point cloud data in the first grid and does not belong to the noise data.

[0091] Step S40: When there is target point cloud in the second grid whose height is related to the height range, determine that the target point cloud is noise data.

[0092] In this embodiment, if laser points are detected in the first grid, it means that there is an object in the actual scene corresponding to the first point cloud data. Due to the characteristics of the laser, it cannot penetrate the existing object. Therefore, in the second grid behind the laser irradiation direction, there cannot be laser points whose height is related to the height range of the first grid, including laser points whose height is within the height range of the first grid or exceeds the height range of the first grid within a certain preset value. Such laser points are target point clouds, and the target point clouds are noise data. Further, if there are a preset number of laser points in the first grid, it is considered that there is an obstacle blocking in the actual operating environment corresponding to the grid. According to the characteristics of the laser, it can be determined that there will be no point cloud data in the second grid whose height is related to the height range of the point cloud data in the first grid.

[0093] Optionally, traverse a preset number of the second grids. If there is target point cloud whose height is within the height range, determine that the target point cloud is noise data.

[0094] The number of second grids can be one or more grids. Traverse a preset number of second grids one by one, and determine the height of each laser point in each grid. If there is a laser point in the second grid whose height is within the height range of the first grid, the laser point is a target point cloud, and the target point cloud data is noise data. Refer to Figure 3, for example, the height range within the first grid 2 is 2 to 6. Traverse the second grids with a preset quantity of 2, namely grid 5 and grid 8. There are laser points A, B, C, D in grid 5, with corresponding heights of 1, 3, 6, 8. Then, laser points B and C are within the height range of 2 to 6. Laser points B and C are the target point clouds and are noise data. Laser points A and D are not within the height range of 2 to 6. Laser points A and D are not the target point clouds. After traversing grid 5, traverse grid 8. There are laser points E and F in grid 8, with corresponding heights of 4 and 5. Laser points E and F are within the height range of 2 to 6. Laser points E and F are the target point clouds and are noise data.

[0095] Optionally, traverse a preset number of the second grids. If the ratio of the number of laser points in the first grid to the number of laser points in the second grid is greater than or equal to a preset ratio, then determine that the point cloud data in the second grid is noise data.

[0096] The number of second grids can be one or more grids. Traverse a preset number of second grids one by one. First, confirm the number of laser points in the first grid, then determine the number of laser points in each second grid, and calculate the ratio of the number of laser points in the first grid to the number of laser points in each second grid. If the ratio is greater than the preset ratio, it means that the number in the second grid is relatively small compared to the number of laser points in the first grid. Then determine that the point cloud data in the second grid corresponding to this ratio is noise data.

[0097] Optionally, calculate the height difference between the first grid and the second grid; if the sum of the height of the laser points in the second grid and the height difference is within the height range, then the laser points are the target point clouds, and determine that the target point clouds are noise data.

[0098] Calculate the height difference of the point cloud data between the first grid and the second grid, and then calculate the sum of the height of each laser point in the second grid and the height difference. If the sum value is within the height range of the point cloud data of the first grid, then the laser points in the second grid corresponding to this sum value are noise data. This takes into account the situation where the first grid and the second grid are not on the same horizontal plane. Only relying on the fact that the laser point height is not within the height range cannot confirm the target point cloud, further improving the recognition rate of crosstalk data.

[0099] Optionally, remove the target point clouds.

[0100] The target point cloud is noise data. Specifically, the target point cloud is crosstalk noise data. Since the point cloud data is used as the main data for navigation, obstacle avoidance, path planning, positioning, etc., and detailed calculation operations will be performed on the point cloud data later, the target point cloud will interfere with the above operations. To reduce unnecessary calculations, improve the work efficiency and accuracy of navigation, obstacle avoidance, path planning, positioning, etc., and reduce the storage space, the target point cloud should be removed.

[0101] Furthermore, considering the actual application scenario, when an object appears in the laser scanning range, the point cloud data corresponding to the first grid will be very dense, and it is not excluded that there will be noise data in the first grid. Therefore, when there is a target point cloud in the second grid whose height is related to the height range in the first grid, it can also be determined that the point cloud data in the first grid is noise data.

[0102] In an alternative implementation of this embodiment, the point cloud data of the lidar is obtained, and the point cloud data is segmented by grids; the first grid and the second grid passed through in sequence by the lidar emission direction are determined; the height range of the point cloud data in the first grid is calculated; when there is a target point cloud in the second grid whose height is related to the height range, it is determined that the target point cloud is noise data. In this way, by segmenting the lidar point cloud data by grids and determining the second grid behind the first grid according to the laser emission direction, since the laser cannot detect the data inside the object, it can be inferred that it is unreasonable for there to be a target point cloud in the second grid whose height is related to the point cloud data height range in the first grid. The possibility of the above situation is crosstalk noise from laser signals emitted by other systems. Such a target point cloud is noise data. Therefore, based on the above embodiments, the recognition rate of crosstalk noise data can be improved, and since it directly processes the lidar point cloud data, it can reduce the dependence on other devices or settings and improve the recognition efficiency of crosstalk noise data.

[0103] In addition, an embodiment of the present invention also proposes a noise data recognition device. The noise data recognition device includes a memory, a processor, and a noise data recognition program stored on the memory and executable on the processor. When the noise data recognition program is executed by the processor, it implements the steps of the noise data recognition method described in each of the above embodiments.

[0104] In addition, an embodiment of the present invention also proposes a computer-readable storage medium. A noise data recognition program is stored on the computer-readable storage medium. When the noise data recognition program is executed by a processor, it implements the steps of the noise data recognition method described in each of the above embodiments.

[0105] It should be noted that in this text, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or system. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or system including such element.

[0106] The serial numbers of the above embodiments of the present invention are only for description and do not represent the superiority or inferiority of the embodiments.

[0107] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium as described above (such as ROM / RAM, magnetic disk, optical disc) and includes several instructions to enable the noise data recognition device to execute the methods described in various embodiments of the present invention.

[0108] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A method for identifying noise data, characterized in that, The noise data recognition method includes the following steps: Obtain the point cloud data of the lidar, and segment the point cloud data through a grid; Determine the first grid and the second grid that the lidar emission direction passes through in sequence, including: traversing the grid along the lidar emission direction, if there is a target grid with the number of laser points greater than or equal to a preset number, then determine the target grid as the first grid; starting from the first grid and with the lidar emission direction as the extension direction, determine the second grid; Calculate the height range of the point cloud data in the first grid; When there is target point cloud in the second grid whose height is related to the height range, determine that the target point cloud is noise data.

2. The noise data recognition method according to claim 1, characterized in that The step of determining the second grid starting from the first grid and with the lidar emission direction as the extension direction includes: Taking the first grid as the starting point and the lidar emission direction as the extension direction, the preset number of grids passed through are respectively used as the second grid.

3. The noise data identification method according to claim 2, wherein The step of determining that the target point cloud is noise data when there is target point cloud in the second grid whose height is related to the height range includes: Traverse a preset number of the second grids, if there is target point cloud with a height within the height range, determine that the target point cloud is noise data.

4. The noise data recognition method according to claim 2, wherein The method further includes: Traverse a preset number of the second grids, if the ratio of the number of laser points in the first grid to the number of laser points in the second grid is greater than or equal to a preset ratio, then determine that the point cloud data in the second grid is noise data.

5. The noise data recognition method according to claim 1, wherein The step of determining that the target point cloud is noise data when there is target point cloud in the second grid whose height is related to the height range includes: Calculate the height difference between the first grid and the second grid; If the sum of the height of the laser points in the second grid and the height difference is within the height range, then the laser points are target point cloud, and determine that the target point cloud is noise data.

6. The noise data recognition method according to claim 1, wherein The step of obtaining the point cloud data of the lidar and segmenting the point cloud data through a grid includes: Obtain the point cloud data of the lidar, and determine the boundary information and grid resolution of the point cloud data; Determine grid parameters according to the boundary information and the grid resolution, and the grid parameters include grid shape and grid size; Segment the point cloud data according to the grid parameters.

7. The noise data recognition method according to claim 1, wherein After determining that the target point cloud is noise data when there is target point cloud in the second grid whose height is related to the height range, it further includes: Remove the target point cloud.

8. A noise data recognition device, characterized in that, The noise data recognition device includes: a memory, a processor, and a noise data recognition program stored on the memory and executable on the processor, and when the noise data recognition program is executed by the processor, it implements the steps of the noise data recognition method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, A noise data recognition program is stored on the computer-readable storage medium, and when the noise data recognition program is executed by the processor, it implements the steps of the noise data recognition method according to any one of claims 1 to 7.

Images