Point cloud noisy point determination method and device, and storage medium

Abstract

The invention provides a point cloud noisy point determination method and device and a storage medium, and the method comprises the steps: determining adjacent point data adjacent to each point data in a point cloud based on the device parameters of a collection device and the three-dimensional coordinates included in each point data, and determining the point cloud noisy point in the point cloud based on the distance. According to the scheme provided by the embodiment of the invention, through the determination of the adjacent point data, the disordered and semantic-free point data in the point cloud is changed into the point data with the relative position relationship, so that the determination of the point cloud noisy points based on the distance becomes possible. Therefore, by adopting the scheme of the embodiment, unstable noisy points such as leaves can be removed at a relatively high accurate removal rate, the point cloud registration effect is effectively improved, map ghosting is eliminated, and the map quality is improved. In addition, the scheme of the embodiment does not need a large number of computing resources and data training, real-time processing can be achieved, and high practicability and feasibility are achieved.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and specifically relates to a point cloud noise determination method, device and storage medium. Background Art

[0002] A vehicle operating in an automated driving mode (e.g., without a human operator) can relieve passengers, particularly the driver, from some driving-related responsibilities. When operating in an automated driving mode, the vehicle can navigate to various locations using onboard sensors, allowing the vehicle to operate with minimal human interaction or in some situations without any passengers.

[0003] Motion planning based on three-dimensional (3D) point cloud maps is a key operation in autonomous driving. Given the high precision and accuracy of LiDAR, accurate 3D point cloud maps are usually generated based on LiDAR. LiDAR collects high-density point cloud data, and performs point cloud matching based on the point cloud data to generate a 3D point cloud map. Currently, in point cloud matching, there are often a large number of noise points caused by unstable factors such as leaves in scenes with a lot of vegetation such as trees. These noise points have a relatively large impact on point cloud registration, which will cause serious ghosting in the final generated 3D point cloud map, greatly affecting the map quality. Summary of the invention

[0004] The present application proposes a point cloud noise determination method, device and storage medium, which can alleviate the technical problem in related technologies that the detection accuracy is not high due to the mismatch between the receptive field and the target object size.

[0005] In a first aspect, a method for determining point cloud noise is provided, comprising:

[0006] Acquire data of each point in the point cloud collected by the collection device, where the collection time of each point in the point cloud is the same;

[0007] Based on the device parameters of the acquisition device and the three-dimensional coordinates included in each of the point data, determine adjacent point data adjacent to each of the point data in the point cloud;

[0008] Based on the distance included in each point data and the distance included in the adjacent point data, the point cloud noise in the point cloud is determined, and the distance included in each point data is the distance between the acquisition device and the detection point corresponding to each point data.

[0009] In some embodiments, based on the device parameters of the acquisition device and the three-dimensional coordinates included in each of the point data, determining adjacent point data adjacent to each of the point data in the point cloud includes:

[0010] Calculate the pixel coordinates of each of the point data based on the device parameters and the three-dimensional coordinates of each of the point data;

[0011] Based on the pixel coordinates of each of the point data, adjacent point data adjacent to each of the point data in the point cloud are determined.

[0012] In some embodiments, calculating the pixel coordinates of each of the point data based on the device parameters and the three-dimensional coordinates of each of the point data includes:

[0013] Based on the horizontal field of view angle, horizontal resolution and the three-dimensional horizontal coordinates and three-dimensional vertical coordinates of each point data included in the device parameters, the pixel horizontal coordinate of each point data is calculated; and based on the vertical field of view angle, vertical resolution and the three-dimensional coordinates of each point data included in the device parameters, the pixel vertical coordinate of each point data is calculated.

[0014] In some embodiments, based on the horizontal field angle and horizontal resolution included in the device parameters and the three-dimensional coordinates of each point data and the three-dimensional coordinates and three-dimensional coordinates included in each point data, calculating the pixel horizontal coordinate of each point data comprises:

[0015] Based on the three-dimensional horizontal coordinate and the three-dimensional vertical coordinate, calculate the horizontal azimuth angle when the acquisition device acquires the data of each point;

[0016] Calculating the difference between the horizontal field of view angle and the horizontal azimuth angle to obtain a horizontal angle difference;

[0017] The quotient of the horizontal angle difference and the horizontal resolution is calculated to obtain the pixel horizontal coordinate of each point data.

[0018] In some embodiments, based on the vertical field angle, the vertical resolution and the three-dimensional coordinates of each of the point data included in the device parameters, calculating the pixel vertical coordinate of each of the point data includes:

[0019] Based on the three-dimensional coordinates, calculate the plane angle between each point data and the horizontal and vertical coordinate planes, where the horizontal and vertical coordinate planes are planes formed by the horizontal and vertical axes of the acquisition device coordinate system;

[0020] Calculating the difference between the absolute value of the vertical field of view angle and the plane angle to obtain a vertical angle difference;

[0021] The quotient of the vertical angle difference and the vertical resolution is calculated to obtain the pixel vertical coordinate of each point data.

[0022] In some embodiments, determining point cloud noise in the point cloud based on the distance included in each of the point data and the distance included in the adjacent point data includes:

[0023] Based on the distance included in each point data and the distance included in the adjacent point data, taking each point data as a cluster center, searching for multiple target point data belonging to the same type as each point data from the adjacent point data to obtain multiple point data sets, each of which corresponds to one type;

[0024] Point data other than the plurality of point data sets in the point cloud are determined to be the point cloud noise.

[0025] In some embodiments, based on the distance included in each point data and the distance included in the adjacent point data, each point data is used as a cluster center, and multiple target point data belonging to the same type as each point data are searched from the adjacent point data to obtain multiple point data sets, including:

[0026] Calculating the distance difference between the distance included in each of the point data and the distance included in the adjacent point data;

[0027] Based on the adjacent point data whose distance difference is less than the distance threshold and each of the point data, each of the point data sets is obtained.

[0028] In a second aspect, a point cloud noise determination device is provided, comprising:

[0029] An acquisition module, used to acquire data of each point in a point cloud acquired by an acquisition device, wherein the acquisition time of each point in the point cloud is the same;

[0030] A first determination module, configured to determine adjacent point data adjacent to each of the point data in the point cloud based on device parameters of the acquisition device and the three-dimensional coordinates included in each of the point data;

[0031] The second determination module is used to determine the point cloud noise in the point cloud based on the distance included in each point data and the distance included in the adjacent point data, and the distance included in each point data is the distance between the acquisition device and the detection point corresponding to each point data.

[0032] According to a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0033] According to a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and the program is executed by a processor to implement the method described in the first aspect.

[0034] The technical solution provided in the embodiments of the present application has at least the following technical effects or advantages:

[0035] In the solution provided by this embodiment, based on the device parameters of the acquisition device and the three-dimensional coordinates included in each point data, the adjacent point data adjacent to each point data in the point cloud are determined, and then on this basis, the point cloud noise in the point cloud is determined based on the distance. The solution of this embodiment transforms the disordered and non-semantic point data in the point cloud into point data with a relative position relationship through the determination of the adjacent point data, thereby making it possible to determine the point cloud noise based on the distance. Therefore, the solution of this embodiment can remove unstable noise points such as leaves with a higher accuracy rate, effectively improve the point cloud registration effect, eliminate map ghosting, and improve the map quality. In addition, the solution of this embodiment does not require a large amount of computing resources and data training, can be processed in real time, and has high practicality and feasibility.

[0036] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] By reading the detailed description of the preferred embodiment below, various other advantages and benefits will become clear to those of ordinary skill in the art. The accompanying drawings are only used for the purpose of illustrating the preferred embodiment and are not considered to be limitations of the present application. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings.

[0038] In the attached picture:

[0039] Figure 1 A flow chart of a method for determining point cloud noise provided by an embodiment of the present application is shown;

[0040] Figure 2 A schematic diagram showing point data of an observation point P provided in an embodiment of the present application in a radar coordinate system is shown;

[0041] Figure 3 A schematic diagram of a tree-lined road collected by the collection equipment;

[0042] Figure 4 This is the mapping result obtained without removing point cloud noise such as leaves;

[0043] Figure 5 The mapping result obtained after removing point cloud noise such as leaves provided in the embodiment of the present application is shown;

[0044] Figure 6 A schematic diagram of the structure of a point cloud noise determination device provided in an embodiment of the present application is shown;

[0045] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present application is shown;

[0046] Figure 8A schematic diagram of a storage medium provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0047] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0048] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by technicians in the field to which this application belongs.

[0049] Embodiments of the present application may involve autonomous driving scenarios, in which motion planning based on three-dimensional (3D) point cloud maps is a key operation in autonomous driving. Currently, three-dimensional point cloud data is obtained by laser radar scanning, and then point cloud matching is performed based on the three-dimensional point cloud data to construct a point cloud map. Currently, in point cloud matching, there are often a large number of noise points caused by unstable factors such as leaves in scenes with a lot of vegetation such as trees. These noise points have a relatively large impact on point cloud registration, which will cause serious ghosting in the final generated three-dimensional point cloud map, greatly affecting the map quality.

[0050] In order to alleviate the impact of noise points caused by unstable factors such as a large number of leaves in scenes with more vegetation such as trees on the quality of three-dimensional point cloud maps, an embodiment of the present application provides a point cloud noise determination method, which determines the adjacent point data adjacent to each point data in the point cloud based on the device parameters of the acquisition device and the three-dimensional coordinates included in each point data, and then on this basis, determines the point cloud noise points in the point cloud based on the distance. The scheme of this embodiment determines the adjacent point data, so that the disordered and non-semantic point data in the point cloud becomes point data with a relative position relationship, thereby making it possible to determine the point cloud noise points based on the distance. Therefore, the scheme of this embodiment can remove unstable noise points such as leaves with a higher accuracy rate, effectively improve the point cloud registration effect, eliminate map ghosting, and improve the map quality. In addition, the scheme of this embodiment does not require a large amount of computing resources and data training, can be processed in real time, and has high practicality and feasibility.

[0051] The sports scenes to which the embodiments of the present application can be applied include, but are not limited to, land traffic scenes, air traffic scenes, and water traffic scenes. In land traffic scenes, the devices to which the collection device is applied may include, but are not limited to, cars, motorcycles, electric vehicles, bicycles, pedestrians, etc. In air traffic scenes, the devices to which the collection device is applied may include, but are not limited to, various aircraft such as airplanes, helicopters, and drones. In water traffic scenes, the devices to which the collection device is applied may include, but are not limited to, various ships, jet skis, water bikes, etc.

[0052] The executor of the technical solution of the embodiment of the present application can be any device that supports point cloud noise determination, including vehicles, ships, aircraft or robots. Such devices can support point cloud noise determination of the embodiment of the present application through an automatic driving system. The automatic driving system can be provided with relevant functional mechanisms such as a point cloud map construction component and a motion control component.

[0053] In the embodiment of the present application, a collection device is provided on the device as the execution subject, and the collection device includes but is not limited to a laser radar, etc. During the movement, the device collects the three-dimensional coordinates of the detection points on the surface of the surrounding objects through the collection device, and determines the point cloud noise based on the three-dimensional coordinates and the device parameters of the collection device.

[0054] A point cloud noise determination method, device, and storage medium proposed according to an embodiment of the present application are described below in conjunction with the accompanying drawings.

[0055] See also Figure 1 , Figure 1 A flow chart of a method for determining point cloud noise provided by an embodiment of the present application is shown, and the method specifically comprises the following steps:

[0056] Step 101: Acquire data of each point in a point cloud collected by a collection device, wherein the collection time of data of each point in the point cloud is the same.

[0057] In this embodiment, the point data includes but is not limited to parameters such as three-dimensional coordinates, the distance between the observation point corresponding to the point data and the acquisition device, etc. The three-dimensional coordinates refer to the coordinates of the observation point on the surface of the object obtained in the coordinate system of the acquisition device.

[0058] In the application, taking the collection device as a laser radar as an example, the transmitting module of the laser radar can emit laser. The laser hits the observation point on the surface of the object and is reflected back and received by the receiver. According to the time difference between the laser radar's transmission and reception, the distance between the laser radar and the observation point can be calculated. Adding the angle information, the three-dimensional coordinates of the observation point can be obtained.

[0059] Step 102: Based on the equipment parameters of the acquisition equipment and the three-dimensional coordinates included in each point data, determine the adjacent point data adjacent to each point data in the point cloud.

[0060] In this embodiment, the device parameters of the acquisition device include but are not limited to parameters such as vertical field of view angle, horizontal field of view angle, vertical resolution, and horizontal resolution.

[0061] In this embodiment, the point cloud includes multiple point data, each of which corresponds to an observation point on the surface of the object. Different observation points on the surface of the object have a relative position relationship, for example, observation points B and C on the surface of the object are located around observation point A, that is, observation point A is adjacent to observation point B and observation point C, observation point D is far away from observation point A and is located around observation point B, that is, observation point D is adjacent to observation point B, and so on.

[0062] In some embodiments, based on the device parameters of the acquisition device and the three-dimensional coordinates included in each point data, determining the adjacent point data adjacent to each point data in the point cloud may include the following steps:

[0063] Calculate the pixel coordinates of each point data based on the device parameters and the three-dimensional coordinates of each point data;

[0064] Based on the pixel coordinates of each point data, adjacent point data adjacent to each point data in the point cloud are determined.

[0065] It should be understood that pixel coordinates are two-dimensional coordinates, which include pixel abscissas and ordinates, wherein the abscissas are used to represent the height position of the pixel in the image, and the ordinates are used to represent the width position of the pixel in the image. Each point data corresponds to a pixel indicated by the pixel coordinates, and the pixel coordinates of the pixel points converted from all the point data in the point cloud constitute the two-dimensional image corresponding to the point cloud.

[0066] In this embodiment, the adjacent pixel points in the image can be found by the pixel coordinates corresponding to each point data, and then the two point data corresponding to the adjacent pixel points are determined as adjacent point data. It should be understood that for a pixel point (a, b) in the image, the coordinates of the pixel points adjacent to the pixel point in the image can be (a+c, b)(ac, b)(a, b+d)(a, bd)(a+c, b+d)(ac, bd), where c is the difference between the pixel horizontal coordinates of two adjacent pixel points with the same image vertical coordinate in the image, and d is the difference between the pixel vertical coordinates of two adjacent pixel points with the same image horizontal coordinate in the image.

[0067] In this embodiment, each point data is converted into a pixel point represented by pixel coordinates, and finally the adjacent point data adjacent to each point data in the point cloud is determined based on the pixel coordinates. The adjacent point data can be determined quickly and accurately through the pixel coordinates, which helps to accurately determine the noise point.

[0068] In this embodiment, the device parameters of the acquisition device and the three-dimensional coordinates included in each point data can be identified through a pre-trained point cloud denoising model, so as to obtain adjacent point data adjacent to each point data.

[0069] In an optional embodiment, the pixel coordinates of each point data are calculated based on the device parameters and the three-dimensional coordinates of each point data, including:

[0070] Based on the equipment parameters including the horizontal field of view angle, horizontal resolution, and the three-dimensional horizontal coordinates and three-dimensional vertical coordinates of each point data, the pixel horizontal coordinate of each point data is calculated; and based on the equipment parameters including the vertical field of view angle, vertical resolution, and the three-dimensional coordinates of each point data, the pixel vertical coordinate of each point data is calculated.

[0071] The pixel horizontal coordinate of each point data is calculated based on the horizontal field angle, horizontal resolution and three-dimensional coordinates of each point data including the three-dimensional horizontal coordinate and three-dimensional vertical coordinate, including:

[0072] Based on the three-dimensional horizontal coordinate and the three-dimensional vertical coordinate, calculate the horizontal azimuth angle when the acquisition device collects data at each point;

[0073] Calculate the difference between the horizontal field of view angle and the horizontal azimuth angle to obtain the horizontal angle difference;

[0074] Calculate the quotient of the horizontal angle difference and the horizontal resolution to obtain the pixel horizontal coordinate of each point data.

[0075] The calculation formula of the horizontal azimuth angle can be:

[0076]

[0077] Among them, β is the horizontal azimuth angle, x is the three-dimensional horizontal coordinate, and y is the three-dimensional vertical coordinate.

[0078] The calculation formula of the pixel horizontal coordinate is:

[0079]

[0080] Among them, μ is the pixel horizontal coordinate, B is the horizontal field angle, r β is the horizontal resolution.

[0081] The vertical coordinate of each point data is calculated based on the vertical field angle, vertical resolution and three-dimensional coordinates of each point data including the device parameters, including:

[0082] Based on the three-dimensional coordinates, the plane angle between each point data and the horizontal and vertical coordinate planes is calculated, and the horizontal and vertical coordinate planes are the planes formed by the horizontal and vertical axes of the acquisition device coordinate system;

[0083] Calculate the difference between the absolute value of the vertical field of view angle and the plane angle to obtain the vertical angle difference;

[0084] Calculate the quotient of the vertical angle difference and the vertical resolution to obtain the pixel vertical coordinate of each point data.

[0085] The calculation formula of the plane angle is:

[0086]

[0087]

[0088] Among them, α is the plane angle, v is the pixel vertical coordinate, A is the vertical field angle, r α is the vertical resolution, x is the three-dimensional horizontal coordinate, y is the three-dimensional vertical coordinate, and z is the three-dimensional vertical coordinate.

[0089] For ease of understanding, the following is given Figure 2 The schematic diagram of the point data of the observation point P in the radar coordinate system is shown.

[0090] Step 103: determine the point cloud noise in the point cloud based on the distance included in each point data and the distance included in the adjacent point data. The distance included in each point data is the distance between the acquisition device and the detection point corresponding to each point data.

[0091] In this embodiment, a cluster segmentation strategy based on regional growth can be adopted, and the point cloud noise in the point cloud can be determined by combining the distance included in each point data and the distance included in the adjacent point data. The cluster segmentation strategy based on regional growth here means that based on the distance included in each point data and the distance included in the adjacent point data, the point data that does not belong to the same type as each data is found in the adjacent point data, and the point data that does not belong to the same type is determined to be a point cloud noise. Then, based on each adjacent point data of the same type, the iterative growth outward continues violently until the iterative growth of all point data is completed.

[0092] In some embodiments, determining point cloud noise in the point cloud based on the distance included in each point data and the distance included in the adjacent point data includes:

[0093] Based on the distance included in each point data and the distance included in the adjacent point data, each point data is used as a cluster center, and multiple target point data belonging to the same type as each point data are searched from the adjacent point data to obtain multiple point data sets, each of which corresponds to one type;

[0094] Point data other than a plurality of point data sets in the point cloud are determined to be point cloud noise.

[0095] The type in this embodiment refers to the type of object to which the observation point represented by the point data belongs, for example, the type may be sky, earth, tree crown, leaves, wall, etc.

[0096] When performing cluster analysis on point data of the same type, this embodiment uses any point data in the image and the adjacent point data adjacent to the arbitrary point data as the minimum clustering unit, and performs cluster analysis based on the distance included in the arbitrary point data and the distance included in each adjacent point data adjacent to the arbitrary point data. It should be understood that in applications, for two point data of the same type, the distances included in the two point data are usually not much different, so the point cloud noise is determined by the iterative growth method of the minimum clustering unit proposed in this embodiment, and the calculation accuracy is high.

[0097] In some embodiments, based on the distance included in each point data and the distance included in the adjacent point data, each point data is used as a cluster center, and multiple target point data belonging to the same type as the point data are searched from the adjacent point data to obtain multiple point data sets, including:

[0098] Calculate the distance difference between the distance included in each point data and the distance included in the adjacent point data;

[0099] Based on the adjacent point data and each point data whose distance difference is less than the distance threshold, a set of each point data is obtained.

[0100] In this embodiment, the distance threshold may be preset artificially based on experience or according to actual needs, and this embodiment does not specifically limit this.

[0101] In this embodiment, after the point cloud noise is determined, the point cloud noise in the point cloud may be removed, and a point cloud map may be generated based on the point cloud after the point cloud noise is removed.

[0102] In an example, a point cloud includes 8 data points. When generating clusters, one data point can be randomly selected from the 8 data points and grown outward with the data point as the center. In specific implementation, it is assumed that the randomly selected data point is data point 1, and the data points adjacent to data point 1 are data point 2 and data point 5. If the distance difference between the distance corresponding to data point 2 and the distance corresponding to data point 1 is less than the distance threshold, and the distance difference between the distance corresponding to data point 5 and the distance corresponding to data point 1 is greater than the distance threshold, it can be determined that the point data corresponding to data point 1 and data point 2 belong to the same point data set, and data point 5 does not belong to the point data set. Next, randomly select another data point from data point 3 to data point 8 as the center to grow outward, and repeat the above steps. Finally, it is found that data point 5 does not belong to any clustered data point set, so it can be determined that data point 5 is a point cloud noise point in the point cloud.

[0103] In order to demonstrate the effect of this embodiment, the following is given: Figure 3-Figure 5 The schematic diagram shown in Figure 3Schematic diagram of a tree-lined road collected by the collection equipment. Figure 4 This is the mapping result obtained without removing point cloud noise such as leaves. Figure 5 This is the result of the map construction after removing point cloud noise such as leaves. Figure 4 and Figure 5 By comparison, it can be found that after removing point cloud noise such as leaves, the mapping result eliminates ghosting and the map quality is greatly improved.

[0104] In the solution provided by this embodiment, based on the device parameters of the acquisition device and the three-dimensional coordinates included in each point data, the adjacent point data adjacent to each point data in the point cloud are determined, and then on this basis, the point cloud noise in the point cloud is determined based on the distance. The solution of this embodiment transforms the disordered and non-semantic point data in the point cloud into point data with a relative position relationship through the determination of the adjacent point data, thereby making it possible to determine the point cloud noise based on the distance. Therefore, the solution of this embodiment can remove unstable noise points such as leaves with a higher accuracy rate, effectively improve the point cloud registration effect, eliminate map ghosting, and improve the map quality. In addition, the solution of this embodiment does not require a large amount of computing resources and data training, can be processed in real time, and has high practicality and feasibility.

[0105] The present application also provides a point cloud noise determination device, which is used to execute the point cloud noise determination method provided in any of the above embodiments. Figure 6 As shown, the device comprises:

[0106] An acquisition module 61 is used to acquire data of each point in the point cloud acquired by an acquisition device, and the acquisition time of each point in the point cloud is the same;

[0107] A first determination module 62, configured to determine adjacent point data adjacent to each point data in the point cloud based on device parameters of the acquisition device and the three-dimensional coordinates included in each point data;

[0108] The second determination module 63 is used to determine the point cloud noise in the point cloud based on the distance included in each point data and the distance included in the adjacent point data. The distance included in each point data is the distance between the acquisition device and the detection point corresponding to each point data.

[0109] Optionally, the first determination module 62 is used to:

[0110] Calculate the pixel coordinates of each point data based on the device parameters and the three-dimensional coordinates of each point data;

[0111] Based on the pixel coordinates of each point data, adjacent point data adjacent to each point data in the point cloud are determined.

[0112] Optionally, the first determination module 62 is used to:

[0113] Based on the equipment parameters including the horizontal field of view angle, horizontal resolution, and the three-dimensional horizontal coordinates and three-dimensional vertical coordinates of each point data, the pixel horizontal coordinate of each point data is calculated; and based on the equipment parameters including the vertical field of view angle, vertical resolution, and the three-dimensional coordinates of each point data, the pixel vertical coordinate of each point data is calculated.

[0114] Optionally, the first determination module 62 is used to:

[0115] Based on the three-dimensional horizontal coordinate and the three-dimensional vertical coordinate, calculate the horizontal azimuth angle when the acquisition device collects data at each point;

[0116] Calculate the difference between the horizontal field of view angle and the horizontal azimuth angle to obtain the horizontal angle difference;

[0117] Calculate the quotient of the horizontal angle difference and the horizontal resolution to obtain the pixel horizontal coordinate of each point data.

[0118] Optionally, the first determination module 62 is used to:

[0119] Based on the three-dimensional coordinates, the plane angle between each point data and the horizontal and vertical coordinate planes is calculated, and the horizontal and vertical coordinate planes are the planes formed by the horizontal and vertical axes of the acquisition device coordinate system;

[0120] Calculate the difference between the absolute value of the vertical field of view angle and the plane angle to obtain the vertical angle difference;

[0121] Calculate the quotient of the vertical angle difference and the vertical resolution to obtain the pixel vertical coordinate of each point data.

[0122] Optionally, the second determining module 63 is used to:

[0123] Based on the distance included in each point data and the distance included in the adjacent point data, each point data is used as a cluster center, and multiple target point data belonging to the same type as each point data are searched from the adjacent point data to obtain multiple point data sets, each of which corresponds to one type;

[0124] Point data other than a plurality of point data sets in the point cloud are determined to be point cloud noise.

[0125] Optionally, the second determining module 63 is used to:

[0126] Calculate the distance difference between the distance included in each point data and the distance included in the adjacent point data;

[0127] Based on the adjacent point data and each point data whose distance difference is less than the distance threshold, a set of each point data is obtained.

[0128] The point cloud noise determination device provided in the embodiment of the present application and the point cloud noise determination method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0129] The present application also provides an electronic device to perform the above point cloud noise determination method. Figure 7 It shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 7 As shown, the electronic device 7 includes: a processor 700, a memory 701, a bus 702 and a communication interface 703. The processor 700, the communication interface 703 and the memory 701 are connected via the bus 702; the memory 701 stores a computer program that can be run on the processor 700, and when the processor 700 runs the computer program, it executes the point cloud noise determination method provided in any of the aforementioned embodiments of the present application.

[0130] The memory 701 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the device network element and at least one other network element is realized through at least one communication interface 703 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0131] The bus 702 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 701 is used to store programs, and the processor 700 executes the program after receiving an execution instruction. The point cloud noise determination method disclosed in any implementation of the embodiment of the present application may be applied to the processor 700, or implemented by the processor 700.

[0132] The processor 700 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 700. The above processor 700 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 readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 701, and the processor 700 reads the information in the memory 701 and completes the steps of the above method in combination with its hardware.

[0133] The electronic device provided in the embodiment of the present application and the point cloud noise determination method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0134] The present application also provides a computer-readable storage medium corresponding to the point cloud noise determination method provided in the above embodiment. Figure 8 The computer-readable storage medium shown is a CD 30 on which a computer program (ie, a program product) is stored. When the computer program is run by a processor, the point cloud noise determination method provided in any of the aforementioned embodiments will be executed.

[0135] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0136] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the point cloud noise determination method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0137] It should be noted that:

[0138] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this description.

[0139] Similarly, it should be understood that in order to streamline the present application and help understand one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be interpreted as reflecting the following schematic diagram: the claimed application requires more features than the features clearly stated in each claim. More specifically, as reflected in the claims below, the inventive aspects are less than all the features of the single embodiment disclosed above. Therefore, the claims following the specific embodiment are hereby expressly incorporated into the specific embodiment, wherein each claim itself serves as a separate embodiment of the present application.

[0140] In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present application and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0141] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for determining point cloud noise, characterized in that: include: Acquire data of each point in the point cloud collected by the collection device, where the collection time of each point in the point cloud is the same; Based on the device parameters of the acquisition device and the three-dimensional coordinates included in each of the point data, determine adjacent point data adjacent to each of the point data in the point cloud; Based on the distance included in each point data and the distance included in the adjacent point data, the point cloud noise in the point cloud is determined, and the distance included in each point data is the distance between the acquisition device and the detection point corresponding to each point data.

2. The method according to claim 1, characterized in that Based on the device parameters of the acquisition device and the three-dimensional coordinates included in each of the point data, determining adjacent point data adjacent to each of the point data in the point cloud, including: Calculate the pixel coordinates of each of the point data based on the device parameters and the three-dimensional coordinates of each of the point data; Based on the pixel coordinates of each of the point data, adjacent point data adjacent to each of the point data in the point cloud are determined.

3. The method according to claim 2, characterized in that Calculating the pixel coordinates of each of the point data based on the device parameters and the three-dimensional coordinates of each of the point data includes: Based on the horizontal field of view angle, horizontal resolution and the three-dimensional horizontal coordinates and three-dimensional vertical coordinates of each point data included in the device parameters, the pixel horizontal coordinate of each point data is calculated; and based on the vertical field of view angle, vertical resolution and the three-dimensional coordinates of each point data included in the device parameters, the pixel vertical coordinate of each point data is calculated.

4. The method according to claim 3, characterized in that Based on the horizontal field angle and horizontal resolution included in the device parameters and the three-dimensional coordinates of each point data including the three-dimensional coordinate and the three-dimensional coordinate, calculating the pixel horizontal coordinate of each point data includes: Based on the three-dimensional horizontal coordinate and the three-dimensional vertical coordinate, calculate the horizontal azimuth angle when the acquisition device acquires the data of each point; Calculating the difference between the horizontal field of view angle and the horizontal azimuth angle to obtain a horizontal angle difference; The quotient of the horizontal angle difference and the horizontal resolution is calculated to obtain the pixel horizontal coordinate of each point data.

5. The method according to claim 3, characterized in that: Calculating the pixel vertical coordinate of each of the point data based on the vertical field angle, the vertical resolution and the three-dimensional coordinate of each of the point data included in the device parameters includes: Based on the three-dimensional coordinates, calculate the plane angle between each point data and the horizontal and vertical coordinate planes, where the horizontal and vertical coordinate planes are planes formed by the horizontal and vertical axes of the acquisition device coordinate system; Calculating the difference between the absolute value of the vertical field of view angle and the plane angle to obtain a vertical angle difference; The quotient of the vertical angle difference and the vertical resolution is calculated to obtain the pixel vertical coordinate of each point data.

6. The method according to any one of claims 1 to 5, characterized in that: Determining point cloud noise points in the point cloud based on the distances included in each of the point data and the distances included in the adjacent point data, comprising: Based on the distance included in each point data and the distance included in the adjacent point data, taking each point data as a cluster center, searching for multiple target point data belonging to the same type as each point data from the adjacent point data to obtain multiple point data sets, each of which corresponds to one type; Point data other than the plurality of point data sets in the point cloud are determined to be the point cloud noise.

7. The method according to claim 6, characterized in that Based on the distances included in each of the point data and the distances included in the adjacent point data, taking each of the point data as a cluster center, searching for multiple target point data belonging to the same type as each of the point data from the adjacent point data to obtain multiple point data sets, including: Calculating the distance difference between the distance included in each of the point data and the distance included in the adjacent point data; Based on the adjacent point data whose distance difference is less than the distance threshold and each of the point data, each of the point data sets is obtained.

8. A point cloud noise determination device, characterized in that: include: An acquisition module, used to acquire data of each point in a point cloud acquired by an acquisition device, wherein the acquisition time of each point in the point cloud is the same; A first determination module, configured to determine adjacent point data adjacent to each of the point data in the point cloud based on device parameters of the acquisition device and the three-dimensional coordinates included in each of the point data; The second determination module is used to determine the point cloud noise in the point cloud based on the distance included in each point data and the distance included in the adjacent point data, and the distance included in each point data is the distance between the acquisition device and the detection point corresponding to each point data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor runs the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method according to any one of claims 1 to 7.

Images