Point cloud data processing methods, devices, electronic equipment and storage media

Abstract

This disclosure relates to a method, apparatus, electronic device, and storage medium for processing point cloud data. The method includes: acquiring point cloud data to be processed; projecting the point cloud data to be processed onto a target raster image to obtain a grayscale image to be processed; determining a target raster containing noise in the grayscale image to be processed based on noise estimation of the grayscale image; and removing noise from the point cloud data to be processed based on the target raster to obtain target point cloud data. By converting the point cloud data to be processed into a grayscale image, the location of noise in the grayscale image can be obtained using relevant noise processing methods, thereby achieving noise processing of the point cloud data and significantly improving the efficiency of point cloud data noise processing.

Description

Technical Field

[0001] This disclosure relates to the field of point cloud technology, and in particular to point cloud data processing methods, apparatus, electronic devices and storage media. Background Technology

[0002] Autonomous driving technology is being widely applied in mining environments. LiDAR systems installed on vehicles can generate large amounts of point cloud data, which can then be used for vehicle localization, mapping, or obstacle detection. However, due to the complexity of mining environments, LiDAR systems are susceptible to various forms of interference, potentially leading to noise in the generated point cloud data, which could severely impact the autonomous driving system.

[0003] In related technologies, the typical approach is to perform final target detection on point cloud data, followed by secondary processing to filter out invalid targets (e.g., deleting dust-related target detection results). However, this imposes a significant computational burden on the backend processing, resulting in low efficiency in point cloud data processing. Summary of the Invention

[0004] This disclosure provides a point cloud data processing method, apparatus, electronic device, and storage medium.

[0005] According to a first aspect of this disclosure, a point cloud data processing method is provided, the method comprising:

[0006] Acquire point cloud data to be processed;

[0007] The point cloud data to be processed is projected onto the target raster image to obtain the grayscale image to be processed;

[0008] Based on the noise estimation of the grayscale image to be processed, the target raster containing noise in the grayscale image to be processed is determined;

[0009] The target point cloud data is obtained by removing noise from the point cloud data to be processed based on the target raster.

[0010] According to a second aspect of this disclosure, a point cloud data processing apparatus is provided, the apparatus comprising:

[0011] The data acquisition module is used to acquire point cloud data to be processed;

[0012] The image acquisition module is used to project the point cloud data to be processed onto the target raster image to obtain the grayscale image to be processed.

[0013] The target raster determination module is used to determine the target raster containing noise in the grayscale image to be processed based on the noise estimation of the grayscale image to be processed;

[0014] The noise reduction module is used to remove noise from the point cloud data to be processed based on the target raster to obtain the target point cloud data.

[0015] According to a third aspect of this disclosure, an electronic device is provided. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described above.

[0016] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the methods described above.

[0017] The point cloud data processing method, apparatus, electronic device, and storage medium provided in this disclosure acquire point cloud data to be processed and project it onto a target raster image to obtain a grayscale image to be processed. Based on noise estimation of the grayscale image to be processed, target raster cells containing noise in the grayscale image to be processed are determined, and noise in the point cloud data to be processed is removed based on the target raster cells to obtain target point cloud data. By converting the point cloud data to be processed into a grayscale image to be processed, the embodiments can utilize relevant noise processing methods in the grayscale image to obtain the position of the target raster cells corresponding to the noise in the grayscale image to be processed, thereby achieving noise processing of the point cloud data to be processed and significantly improving the efficiency of point cloud data noise processing. Attached Figure Description

[0018] Further details, features, and advantages of this disclosure are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0019] Figure 1 A flowchart of point cloud data processing provided as an exemplary embodiment of this disclosure;

[0020] Figure 2 A schematic block diagram of the functional modules of a point cloud data processing apparatus provided for an exemplary embodiment of the present disclosure;

[0021] Figure 3 A structural block diagram of an electronic device provided as an exemplary embodiment of this disclosure;

[0022] Figure 4 A block diagram of a computer system provided for an exemplary embodiment of this disclosure. Detailed Implementation

[0023] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0024] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0025] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0026] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0027] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0028] In open-pit mines and other mining scenarios, the working area constantly changes as excavators dig and unmanned transport vehicles transport goods. To enable better positioning of these unmanned transport vehicles, point cloud acquisition devices such as LiDAR can be installed on them or other vehicles to obtain point cloud data. Since these vehicles can continuously travel back and forth within the open-pit mine, they can acquire real-time point cloud data of the mining environment. Furthermore, they can update high-precision maps in real-time or promptly based on this data, allowing the unmanned vehicles to navigate according to the latest high-precision maps, thereby improving driving and transportation efficiency.

[0029] Because the working area in open-pit mines is complex, the point cloud data obtained by LiDAR installed on vehicles or drones may be affected by factors such as road dust or rain and snow. This can lead to noise in the point cloud data, which in turn affects the accuracy of high-precision maps created or updated based on the point cloud data. As a result, unmanned vehicles such as autonomous transport vehicles may experience positioning errors or incorrect changes in their driving routes when driving on these high-precision maps, which may lead to traffic accidents or incorrect driving routes.

[0030] Therefore, it is necessary to remove noise from point cloud data in a timely manner to improve the accuracy of high-precision maps. However, if post-processing is used, the processing efficiency will be low due to the large amount of computation required on the backend. Furthermore, the long processing time due to the large amount of computation required on the backend will prevent the high-precision map from being updated in a timely manner.

[0031] In order to improve the processing efficiency of point cloud data and thus enable timely updates of high-precision maps, in the embodiments provided in this disclosure, noise in the obtained point cloud data can be processed in a timely or even real-time manner to remove noise from the point cloud data, and the high-precision map can be updated in a timely manner based on the processed point cloud data.

[0032] In this embodiment, point cloud data generated by lidar can be received. The working principle of lidar is very similar to that of radar. Using laser as a signal source, the pulsed laser emitted by the laser hits mines, trees, roads, bridges or buildings on the ground, causing scattering. Some of the light waves are reflected to the lidar receiver. Based on the principle of laser ranging, the distance from the lidar to the target point is calculated. The pulsed laser continuously scans the target object, and data of all target points on the target object can be obtained, thus obtaining point cloud data.

[0033] Specifically, taking a single-line lidar as an example, it calculates the precise distance to the observed object by emitting and receiving laser beams. This can be achieved through light speed ranging, specifically using the Time of Flight (TOF) method. First, the lidar emits a laser pulse, and a timer records the emission time. The pulse is reflected by the object and received by the receiver, which also records the reception time. The distance between the lidar and the observed object is obtained by multiplying the time difference between the transmission and reception times by the speed of light and dividing by 2.

[0034] The distance obtained through the Time-of-Flight (TOF) time difference is used to establish a two-dimensional polar coordinate system by scanning the horizontal rotation angle. Then, three-dimensional height information is obtained by acquiring different pitch angles. For example, a 64-line lidar array has 64 laser transmitters and receivers. Each transmission and reception results in a vertical column of 64 points. During acquisition, this column of transmitters and receivers rotates horizontally, acquiring multiple columns of laser points in one 360° rotation. Because the 64 transmitters emit at different angles, they have maximum and minimum horizontal angles. Adjacent transmitters have a certain vertical angular resolution (e.g., 0.4 degrees) and a certain horizontal angular resolution (e.g., 0.08 degrees). That is, the vertical resolution is 0.4°, and the horizontal resolution is 0.08°. Since the vertical spacing is generally much larger than the horizontal spacing, the obtained point cloud data typically presents a concentric, multi-ring shape.

[0035] In this embodiment, azimuth and pitch angles describe the position of an object in space relative to the observer. Related to azimuth and pitch angles is angular resolution, which can be divided into horizontal resolution and vertical resolution, representing the angle between two adjacent points. Horizontal resolution refers to the minimum interval in degrees between two scan lines in the horizontal direction. Since the frequency of the laser beam emitted per second is fixed, the faster the rotation speed, the larger the interval between scan lines in the horizontal direction, resulting in a larger horizontal angular resolution and a larger interval in degrees between the two lines. Vertical resolution refers to the interval in degrees between two scan lines in the vertical direction.

[0036] Therefore, after obtaining point cloud data from sensors such as LiDAR, the point cloud data can be preprocessed. This preprocessing can include outlier removal, filtering, or downsampling to obtain clearer and denser point cloud data. In this embodiment, the processed point cloud data can be used as the point cloud data to be processed. The azimuth and elevation angles corresponding to the point cloud data can be obtained. For example, if the azimuth is 0° to 360°, the horizontal resolution is 0.1°, the elevation angle is -15° to 15°, and the vertical resolution is 0.2°, then based on the azimuth and elevation angles and their corresponding horizontal and vertical resolutions, the data can be divided into 3600 grids in the horizontal direction and 150 grids in the vertical direction, resulting in a 3600*150 grid image. Since the point cloud data can include up to 3600*150 points, each point in the point cloud data can be projected onto the corresponding grid in the grid image. Since each point in the point cloud data carries distance information, which is the distance between the sensor and the object, each grid in the projected grid image can be regarded as a pixel. Each pixel has a corresponding grayscale value, which is the distance between the sensor and the object.

[0037] Therefore, the three projected images can be viewed as grayscale images with image grayscale values. Thus, in this embodiment, noise processing of the point cloud to be processed can be converted into noise processing of the grayscale image, and noise processing of the point cloud data can be viewed as noise processing of the grayscale image.

[0038] In this embodiment, for example, when the LiDAR is installed above a vehicle to acquire point cloud data around the vehicle, the LiDAR can scan the ground and objects above it while the vehicle is moving, due to its pitch angle. This allows it to collect corresponding point cloud data. For instance, if a car appears 5 meters in front of the LiDAR, it will collect point cloud data corresponding to that car. The points in this point cloud data carry distance information, representing the distance between the LiDAR and the car, which should be around 5 meters. If one or more points in the point cloud data contain distances much smaller than 5 meters, such as 3 meters, this indicates noise in the point cloud data. This noise might be caused by dust, rain, or snow between the LiDAR and the car. In this case, high noise needs to be removed from the point cloud data. This embodiment is merely illustrative and is not limited to this.

[0039] Therefore, the grayscale values ​​of the obtained grayscale image should also be within a preset range; values ​​outside this range may be noise. In this embodiment, noise is estimated by performing noise estimation on the grayscale image to estimate the probability of noise, thereby removing noise from the grayscale image.

[0040] In this embodiment, since the correspondence between point cloud data and grayscale image has been established, after removing noise from the grayscale image, the processed grayscale image can be converted into point cloud data through back projection. The converted point cloud data is the point cloud data after noise removal.

[0041] In addition, the embodiment can also obtain the location of noise in the grayscale image and determine the target point corresponding to the noise in the point cloud data. By removing the target point, the point cloud after noise removal can be obtained.

[0042] In the embodiments, each sensor—camera, LiDAR, or millimeter-wave radar—has its own corresponding coordinate system; that is, all data generated by the sensors is based on the sensor's own coordinate system. However, an autonomous vehicle often has many sensors installed on its body. To facilitate algorithm research and testing, the data obtained by each sensor can be converted to the vehicle's coordinate system, i.e., the base coordinate system. This process can be regarded as the external parameter calibration of the sensors. For example, LiDAR is installed on a moving platform. By using GNSS (Global Navigation Satellite System) and an inertial measurement unit (IMU) in conjunction, the LiDAR measurement points can be converted from relative coordinate systems to position points in absolute coordinate systems, thus enabling their application in different systems.

[0043] Based on the above embodiments, in another embodiment provided in this disclosure, a point cloud data processing method is also provided, such as... Figure 1 As shown, the method may include the following steps:

[0044] In step S110, the point cloud data to be processed is acquired.

[0045] In this embodiment, the raw point cloud data acquired by point cloud data acquisition devices such as lidar can be preprocessed, for example, by removing outliers, filtering, or downsampling, to obtain clearer and denser point cloud data. The preprocessed point cloud data can then be used as the point cloud data to be processed. Since the point cloud data to be processed may still contain noise, further processing is required to remove the noise.

[0046] In step S120, the point cloud data to be processed is projected onto the target raster image to obtain the grayscale image to be processed.

[0047] In this embodiment, the azimuth and elevation angles, as well as the angular resolution, corresponding to the point cloud data to be processed can be obtained. A target raster image can be established based on these azimuth, elevation, and angular resolutions. The size of the target raster image is determined by the range corresponding to the azimuth and elevation angles, respectively. The raster density of the target raster image corresponds to the angular resolution. For example, the target raster image can be a 3600*150 raster image established in the above embodiment. This allows each point in the point cloud data to be processed to be projected onto the target raster image. Since the point cloud data to be processed carries distance information—the distance between each point in the point cloud data and the measured object—this projection ensures that the raster in the target raster image contains the distance information of the points in the point cloud data. Thus, the projected target raster image is a grayscale image containing grayscale values. Since the grayscale value of a pixel in a grayscale image ranges from 0 to 255, and since the distance between each point in the point cloud data to be processed and the object being measured varies, the distance information can be normalized to between 0 and 255 to obtain the corresponding grayscale image to be processed.

[0048] In this embodiment, the size of the target raster image is determined based on the azimuth and pitch angles of the point cloud data to be processed, and the points in the point cloud data to be processed correspond to the gratings in the target raster image.

[0049] In step S130, based on the noise estimation of the grayscale image to be processed, the target raster containing noise in the grayscale image to be processed is determined.

[0050] In the embodiment, by converting the point cloud data to be processed into the corresponding grayscale image to be processed, the noise in the grayscale image to be processed can be determined by, for example, quantile noise estimation, histogram noise estimation or minimum noise estimation, i.e., which grids in the grayscale image to be processed the noise is located in, so as to process the noise.

[0051] In one embodiment, at least two noise estimation methods can be used to estimate the noise in the grayscale image to be processed, resulting in at least two noise estimation results. For example, histogram noise estimation and minimum noise estimation methods can be used to obtain histogram noise estimation and minimum noise estimation results, respectively. Then, a final noise estimation result is determined based on these at least two noise estimation results, and subsequently, the target raster is determined based on this final noise estimation result. The final noise estimation result is the intersection of at least two noise estimation results. That is, only the portion that is determined to be noise by all noise estimation methods is considered noise, thereby improving the accuracy of noise estimation.

[0052] In another embodiment, at least two noise estimation methods can be used to identify target gratings containing noise in the grayscale image to be processed, resulting in multiple target grating clusters (one target grating cluster corresponds to a target grating containing noise obtained by one noise estimation method); then, the intersection of the multiple target grating clusters is determined as the final identified target gratings containing noise in the grayscale image to be processed. That is, only the portion that is determined to be noise by all noise estimation methods is considered noise, thereby improving the accuracy of noise estimation.

[0053] In step S140, noise in the point cloud data to be processed is removed based on the target raster to obtain the target point cloud data.

[0054] In this embodiment, the grayscale values ​​in the noisy target raster can be deleted, and then the processed grayscale image to be processed can be converted into the corresponding point cloud data to achieve noise processing of the point cloud data to be processed. Alternatively, when the noisy target raster is identified, the target points in the point cloud data to be processed can be obtained and removed to achieve the purpose of denoising the point cloud data to be processed.

[0055] In one embodiment, the target raster locations for which grayscale values ​​are to be deleted are marked, and the marked target raster locations corresponding to multiple point cloud data to be processed are stored. Using the stored target raster locations, target raster locations whose frequency reaches a frequency threshold are obtained and used as reference raster locations. Based on the reference raster locations, it is determined whether the LiDAR has any dead pixels caused by its own inherent characteristics. This embodiment can assist in the identification of LiDAR malfunctions or dead pixels during point cloud data processing.

[0056] The point cloud data processing method provided in this disclosure acquires point cloud data to be processed and projects it onto a target raster image to obtain a grayscale image to be processed. Based on noise estimation of the grayscale image to be processed, a target raster containing noise in the grayscale image is determined, and noise in the point cloud data to be processed is removed based on the target raster to obtain the target point cloud data. By converting the point cloud data to be processed into a grayscale image, the method can utilize relevant noise processing techniques in the grayscale image to obtain the position of the target raster corresponding to the noise in the grayscale image to be processed, thereby achieving noise processing of the point cloud data and significantly improving the efficiency of point cloud data noise processing.

[0057] Based on the above embodiments, in another embodiment provided in this disclosure, step S140 may further include the following steps:

[0058] Step S141: Remove noise from the grayscale image to be processed based on the target raster to obtain the processed grayscale image.

[0059] Step S142: Convert the processed grayscale image into target point cloud data.

[0060] In this embodiment, after noise estimation of the grayscale image to be processed, the location of the noise in the grayscale image can be determined, i.e., which grids the noise is located in. The target grid to which the noise belongs is identified, and the grayscale values ​​contained in the target grid are removed, thus achieving the purpose of noise processing of the grayscale image to be processed. The processed grayscale image is then converted into target point cloud data. Since the grayscale values ​​contained in the target grid have been removed, the target point cloud data obtained when converting the grayscale image to target point cloud data will not contain the points corresponding to the target grid, thereby achieving noise processing of the point cloud data to be processed.

[0061] It should be noted that, in the process of removing noise from the point cloud data to be processed based on the target grid to obtain the target point cloud data, the embodiments of this disclosure can remove the points corresponding to the noise in the point cloud data to be processed, and can also process the points corresponding to the noise in the point cloud data to be processed. For example, by adjusting the distance information of the points corresponding to the noise in the point cloud data to be processed (which can be simply referred to as noise points), specifically, the distance information corresponding to the noise points can be adjusted based on other points near the noise points, so that the distance information corresponding to the noise points is more consistent with the distance information corresponding to other points around them, thereby achieving the purpose of removing noise from the point cloud data to be processed.

[0062] In one embodiment, when the noise point corresponds to an edge point in the point cloud data to be processed, the distance information corresponding to that noise point is adjusted to match the distance information of other points in its vicinity. For example, the distance information is adjusted to be consistent with or smoothly change with the distance information of surrounding points. When the noise point corresponds to a non-edge point in the point cloud data to be processed, that point can be removed. This embodiment allows for focused repair of edge noise points and direct removal of non-edge noise points, improving data processing efficiency while minimizing edge errors.

[0063] Based on the above embodiments, in another embodiment provided in this disclosure, step S140 may further include the following steps:

[0064] Step S143: Obtain the target point in the point cloud data to be processed corresponding to the target raster.

[0065] Step S144: Remove the target point from the point cloud data to be processed to obtain the target point cloud data.

[0066] After estimating the noise in the grayscale image to be processed, the location of the noise in the grayscale image to be processed can be determined, that is, in which grids the noise is located, and the target grid to which the noise belongs can be determined. Since there is a correspondence between the grids in the target grid image and the points in the point cloud data to be processed, when it is determined that the target grid contains noise, the points in the point cloud data to be processed can be removed, thereby realizing the noise processing of the point cloud data to be processed and greatly improving the processing efficiency of point cloud data.

[0067] Based on the above embodiments, in another embodiment provided in this disclosure, step S120 may further include the following steps:

[0068] Step S121: Obtain the distance information carried by each point in the point cloud data to be processed. This distance information represents the distance between the point cloud collector and the target object.

[0069] Step S122: Determine the grayscale value of the corresponding grid in the target grid image based on the distance information.

[0070] In this embodiment, each point in the point cloud data to be processed includes distance information between the point cloud collector and the target object. Therefore, after projecting the point cloud data to be processed onto the target grid, the distance in this distance information can be used as the grayscale value of the corresponding grid, thus obtaining the corresponding grayscale image to be processed. Since the grayscale value of a grayscale image is generally in the range of 0 to 255, when the distance fluctuation range is relatively large, the distance information corresponding to each point in the point cloud data to be processed can be normalized to the range of 0 to 255, so that noise processing can be performed on the obtained grayscale image to achieve the purpose of noise processing of the point cloud data to be processed.

[0071] By dividing each function into corresponding functional modules, this disclosure provides a point cloud data processing device, which can be a server or a chip applied to a server. Figure 2 This is a schematic block diagram of the functional modules of a point cloud data processing apparatus provided for an exemplary embodiment of this disclosure. Figure 2 As shown, the point cloud data processing device includes:

[0072] Data acquisition module 10 is used to acquire point cloud data to be processed;

[0073] Projection module 20 is used to project the point cloud data to be processed onto the target raster image to obtain the grayscale image to be processed;

[0074] The target raster determination module 30 is used to determine the target raster containing noise in the grayscale image to be processed based on the noise estimation of the grayscale image to be processed.

[0075] The noise reduction module 40 is used to remove noise from the point cloud data to be processed based on the target raster to obtain the target point cloud data.

[0076] In another embodiment provided in this disclosure, the noise reduction module is specifically used for:

[0077] Based on the target raster, noise is removed from the grayscale image to be processed to obtain the processed grayscale image;

[0078] The processed grayscale image is converted into target point cloud data.

[0079] In another embodiment provided in this disclosure, the noise reduction module is specifically used for:

[0080] Obtain the target point in the point cloud data to be processed corresponding to the target raster;

[0081] The target point is removed from the point cloud data to be processed to obtain the target point cloud data.

[0082] In another embodiment provided in this disclosure, the size of the target raster image is determined based on the azimuth and pitch angles of the point cloud data to be processed, and the points in the point cloud data to be processed correspond to the gratings in the target raster image.

[0083] In another embodiment provided in this disclosure, the projection module is specifically used for:

[0084] Obtain the distance information carried by each point in the point cloud data to be processed, wherein the distance information represents the distance between the point cloud collector and the target object;

[0085] The grayscale value of the corresponding grid in the target grid image is determined based on the distance information.

[0086] In yet another embodiment provided in this disclosure, the apparatus further includes:

[0087] The preprocessing module is used to preprocess the raw point cloud data to obtain the point cloud data to be processed; the preprocessing includes removing outliers, filtering, or downsampling.

[0088] In another embodiment provided in this disclosure, the noise reduction module is further configured to:

[0089] Remove the grayscale value corresponding to the target raster.

[0090] The point cloud data processing apparatus provided in this disclosure acquires point cloud data to be processed and projects it onto a target raster image to obtain a grayscale image to be processed. Based on noise estimation of the grayscale image to be processed, a target raster containing noise in the grayscale image is determined, and noise in the point cloud data to be processed is removed based on the target raster to obtain the target point cloud data. By converting the point cloud data to be processed into a grayscale image, the apparatus can utilize relevant noise processing methods in the grayscale image to obtain the position of the target raster corresponding to the noise in the grayscale image to be processed, thereby achieving noise processing of the point cloud data to be processed and significantly improving the efficiency of point cloud data noise processing.

[0091] This disclosure also provides an electronic device, including: at least one processor; a memory for storing processor-executable instructions; wherein the at least one processor is configured to execute the instructions to implement the methods disclosed in this disclosure.

[0092] Figure 3 This is a schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this disclosure. For example... Figure 3 As shown, the electronic device 1800 includes at least one processor 1801 and a memory 1802 coupled to the processor 1801. The processor 1801 can perform the corresponding steps in the methods disclosed in the embodiments of this disclosure.

[0093] The processor 1801 described above can also be called a central processing unit (CPU), which can be an integrated circuit chip with signal processing capabilities. Each step in the method disclosed in this embodiment can be implemented by the integrated logic circuitry in the processor 1801 or by software instructions. The processor 1801 can be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this embodiment can be directly implemented by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software modules can be located in the memory 1802, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor 1801 reads information from the memory 1802 and, in conjunction with its hardware, completes the steps of the method described above.

[0094] Furthermore, various operations / processes according to this disclosure, implemented via software and / or firmware, can be transmitted from a storage medium or network to a computer system with a dedicated hardware architecture, such as... Figure 4 The computer system 1900 shown is equipped with the programs that constitute the software. When various programs are installed, the computer system is able to perform various functions, including those described above. Figure 4 A block diagram of a computer system provided for an exemplary embodiment of this disclosure.

[0095] Computer System 1900 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0096] like Figure 4 As shown, the computer system 1900 includes a computing unit 1901, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 1902 or a computer program loaded from a storage unit 1908 into a random access memory (RAM) 1903. The RAM 1903 may also store various programs and data required for the operation of the computer system 1900. The computing unit 1901, ROM 1902, and RAM 1903 are interconnected via a bus 1904. An input / output (I / O) interface 1905 is also connected to the bus 1904.

[0097] Multiple components in computer system 1900 are connected to I / O interface 1905, including: input unit 1906, output unit 1907, storage unit 1908, and communication unit 1909. Input unit 1906 can be any type of device capable of inputting information into computer system 1900. Input unit 1906 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of the electronic device. Output unit 1907 can be any type of device capable of presenting information and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 1908 may include, but is not limited to, hard disks and optical disks. Communication unit 1909 allows computer system 1900 to exchange information / data with other devices via a network such as the Internet, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth™ devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0098] The computing unit 1901 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1901 performs the various methods and processes described above. For example, in some embodiments, the methods disclosed in this disclosure can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1908. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 1900 via ROM 1902 and / or communication unit 1909. In some embodiments, the computing unit 1901 can be configured to perform the methods disclosed in this disclosure by any other suitable means (e.g., by means of firmware).

[0099] This disclosure also provides a computer-readable storage medium, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is able to perform the methods disclosed in this disclosure.

[0100] The computer-readable storage medium in this disclosure can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. The aforementioned computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specifically, the aforementioned computer-readable storage medium may include electrical connections based on one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0101] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0102] This disclosure also provides a computer program product, including a computer program, wherein the computer program, when executed by a processor, implements the methods disclosed in the embodiments of this disclosure.

[0103] In embodiments of this disclosure, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or it can be connected to an external computer.

[0104] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0105] The modules, components, or units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules, components, or units do not necessarily constitute a limitation on the module, component, or unit itself.

[0106] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that can be used include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0107] The above description is merely an embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0108] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A point cloud data processing method, characterized in that, The method includes: Acquire point cloud data to be processed; The point cloud data to be processed is projected onto the target raster image to obtain the grayscale image to be processed; wherein, the points in the point cloud data to be processed correspond to the rasters in the target raster image; Based on noise estimation of the grayscale image to be processed, target grids containing noise in the grayscale image to be processed are identified; wherein, the grids in the grayscale image to be processed correspond to grayscale values, and the grayscale values ​​are determined based on the distance between the sensor and the object; The target point cloud data is obtained by removing noise from the point cloud data to be processed based on the target raster.

2. The method according to claim 1, characterized in that, The step of removing noise from the point cloud data to be processed based on the target raster to obtain target point cloud data includes: Based on the target raster, noise is removed from the grayscale image to be processed to obtain the processed grayscale image; The processed grayscale image is converted into target point cloud data.

3. The method according to claim 1, characterized in that, The step of removing noise from the point cloud data to be processed based on the target raster to obtain target point cloud data includes: Obtain the target point in the point cloud data to be processed corresponding to the target raster; The target point is removed from the point cloud data to be processed to obtain the target point cloud data.

4. The method according to claim 1, characterized in that, The size of the target raster image is determined based on the azimuth and elevation angles of the point cloud data to be processed.

5. The method according to claim 1, characterized in that, The step of projecting the point cloud data to be processed onto the target raster image includes: Obtain the distance information carried by each point in the point cloud data to be processed, wherein the distance information represents the distance between the point cloud collector and the target object; The grayscale value of the corresponding grid in the target grid image is determined based on the distance information.

6. The method according to claim 1, characterized in that, The method further includes: The original point cloud data is preprocessed to obtain the point cloud data to be processed; the preprocessing includes removing outliers, filtering, or downsampling.

7. The method according to claim 2, characterized in that, The step of removing noise from the grayscale image to be processed based on the target raster includes: Remove the grayscale value corresponding to the target raster.

8. A point cloud data processing device, characterized in that, The device includes: The data acquisition module is used to acquire point cloud data to be processed; The image acquisition module is used to project the point cloud data to be processed onto the target raster image to obtain the grayscale image to be processed; wherein, the points in the point cloud data to be processed correspond to the gratings in the target raster image; The target grid determination module is used to determine the target grids containing noise in the grayscale image to be processed based on noise estimation of the grayscale image to be processed; wherein, the grids in the grayscale image to be processed correspond to grayscale values, and the grayscale values ​​are determined based on the distance between the sensor and the object; The noise reduction module is used to remove noise from the point cloud data to be processed based on the target raster to obtain the target point cloud data.

9. An electronic device, characterized in that, include: At least one processor; Memory for storing the at least one processor-executable instruction; The at least one processor is configured to execute the instructions to implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the method as described in any one of claims 1-7.

Images