Point cloud data denoising method, device and equipment and computer readable storage medium

Abstract

The application discloses a point cloud data denoising method, device and equipment and a computer readable storage medium, and belongs to the technical field of automatic driving. The method comprises the following steps: acquiring initial point cloud data, wherein the initial point cloud data comprises feature information of a plurality of feature points obtained by emitting laser pulses in multiple directions by a laser radar to a target object, the feature information comprises laser emission time, echo receiving time and pulse width information, the laser emission time of any feature point is the time of emitting a laser pulse to the feature point by the laser radar, and the echo receiving time of any feature point is the time of receiving an echo reflected by the feature point by the laser radar; determining noise points in the plurality of feature points according to the laser emission time, the echo receiving time and the pulse width information of the plurality of feature points; and removing the feature information of the noise points from the initial point cloud data to obtain target point cloud data. The accuracy of the target point cloud data obtained by the method is higher.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a point cloud data denoising method, apparatus, device, and computer-readable storage medium. Background Technology

[0002] With the continuous development of autonomous driving technology, more and more vehicles support autonomous driving. LiDAR is the main device in autonomous driving systems used to acquire information about the external environment. The point cloud data obtained by LiDAR scanning the target object is used to determine the shape of the target object, and then guides the autonomous vehicle to drive based on the shape of the target object.

[0003] Due to environmental factors such as rain, snow, and sandstorms, the point cloud data acquired by LiDAR may contain some noise points, resulting in lower accuracy of the acquired point cloud data. Consequently, the accuracy of the shape of the target object determined based on the point cloud data is also lower, further reducing the driving safety of autonomous vehicles. Summary of the Invention

[0004] This application provides a point cloud data denoising method, apparatus, device, and computer-readable storage medium, which can be used to solve the problem of low accuracy of acquired point cloud data in related technologies. The technical solution is as follows:

[0005] On one hand, embodiments of this application provide a point cloud data denoising method, the method comprising:

[0006] Acquire initial point cloud data, which includes feature information of multiple feature points obtained by the lidar emitting laser pulses toward the target object in multiple directions. Each direction corresponds to at least one feature point. The feature information includes laser emission time, echo reception time, and pulse width information. The laser emission time of any feature point is the time when the lidar emits a laser pulse toward the feature point. The echo reception time of any feature point is the time when the lidar receives the echo reflected from the feature point. The pulse width information of any feature point indicates the pulse width of the echo reflected from the feature point.

[0007] Based on the laser emission time, echo reception time, and pulse width information of the multiple feature points, noise points are determined among the multiple feature points;

[0008] The feature information of the noise points is removed from the initial point cloud data to obtain target point cloud data, which is used to determine the shape of the target object.

[0009] In one possible implementation, determining noise points among the plurality of feature points based on the laser emission time, echo reception time, and pulse width information of the plurality of feature points includes:

[0010] Based on the laser emission time and echo reception time of each feature point, the distance information of each feature point is determined, and the distance information of any feature point indicates the distance between the feature point and the lidar.

[0011] Based on the distance information and pulse width information of each feature point, noise points are determined among the multiple feature points.

[0012] In one possible implementation, determining noise points among the plurality of feature points based on the distance information and pulse width information of each feature point includes:

[0013] A reference point is determined from the plurality of feature points, wherein the distance information of the reference point is less than a distance threshold.

[0014] Based on the existence of candidate points among the reference points whose pulse width information is greater than the pulse width threshold, the candidate points are used as the noise points.

[0015] In one possible implementation, the method further includes:

[0016] Based on the distance information and pulse width information of the target point, the reflected energy information of the target point is determined. The reflected energy information of the target point indicates the energy reflected by the target point. The target point is a feature point among the reference points other than the candidate points.

[0017] Based on the fact that the reflected energy information of the target point is less than the reflected energy threshold, the target point is designated as the noise point.

[0018] In one possible implementation, determining the reflected energy information of the target point based on the distance information and pulse width information of the target point includes:

[0019] The product of the distance information and pulse width information of the target point is used as the reflected energy information of the target point.

[0020] In one possible implementation, determining noise points among the plurality of feature points based on the distance information and pulse width information of each feature point includes:

[0021] A reference point is determined from the plurality of feature points, wherein the distance information of the reference point is less than a distance threshold.

[0022] Based on the distance information and pulse width information of each reference point, the reflected energy information of each reference point is determined, and the reflected energy information of any reference point indicates the energy reflected by any reference point.

[0023] The reference points whose reflected energy information is less than the reflected energy threshold are designated as the noise points.

[0024] In one possible implementation, determining the distance information of each feature point based on the laser emission time and echo reception time of each feature point includes:

[0025] For any one of the plurality of feature points, the laser flight time of any one feature point is determined based on the laser emission time and echo reception time of that feature point.

[0026] The distance information of any feature point is determined based on the laser flight speed and the laser flight duration of any feature point.

[0027] On the other hand, embodiments of this application provide a point cloud data denoising device, the device comprising:

[0028] An acquisition module is used to acquire initial point cloud data, which includes feature information of multiple feature points obtained by the lidar emitting laser pulses toward the target object in multiple directions. Each direction corresponds to at least one feature point. The feature information includes laser emission time, echo reception time, and pulse width information. The laser emission time of any feature point is the time when the lidar emits a laser pulse toward the feature point. The echo reception time of any feature point is the time when the lidar receives the echo reflected from the feature point. The pulse width information of any feature point indicates the pulse width of the echo reflected from the feature point.

[0029] The determination module is used to determine noise points among the multiple feature points based on the laser emission time, echo reception time, and pulse width information of the multiple feature points;

[0030] The noise removal module is used to remove the feature information of the noise points from the initial point cloud data to obtain target point cloud data, which is used to determine the shape of the target object.

[0031] In one possible implementation, the determining module is configured to determine the distance information of each feature point based on the laser emission time and echo reception time of each feature point, wherein the distance information of any feature point indicates the distance between the feature point and the lidar; and to determine noise points among the plurality of feature points based on the distance information and pulse width information of each feature point.

[0032] In one possible implementation, the determining module is configured to determine a reference point among the plurality of feature points, wherein the distance information of the reference point is less than a distance threshold; and based on the existence of candidate points among the reference points whose pulse width information is greater than the pulse width threshold, the candidate points are used as the noise points.

[0033] In one possible implementation, the determining module is further configured to determine the reflected energy information of the target point based on the distance information and pulse width information of the target point, wherein the reflected energy information of the target point indicates the energy reflected by the target point, and the target point is a feature point in the reference points other than the candidate points; and based on the fact that the reflected energy information of the target point is less than the reflected energy threshold, the target point is designated as the noise point.

[0034] In one possible implementation, the determining module is configured to use the product of the distance information and the pulse width information of the target point as the reflected energy information of the target point.

[0035] In one possible implementation, the determining module is configured to determine reference points among the plurality of feature points, wherein the distance information of the reference points is less than a distance threshold; determine the reflection energy information of each reference point based on the distance information and pulse width information of each reference point, wherein the reflection energy information of any reference point indicates the energy reflected by any reference point; and designate reference points whose reflection energy information is less than the reflection energy threshold as noise points.

[0036] In one possible implementation, the determining module is configured to, for any one of the plurality of feature points, determine the laser flight duration of the any one feature point based on the laser emission time and echo reception time of the any one feature point; and determine the distance information of the any one feature point based on the laser flight speed and the laser flight duration of the any one feature point.

[0037] On the other hand, embodiments of this application provide a computer device, the computer device including a processor and a memory, the memory storing at least one piece of program code, the at least one piece of program code being loaded and executed by the processor, so that the computer device implements any of the point cloud data denoising methods described above.

[0038] On the other hand, a computer-readable storage medium is also provided, wherein at least one piece of program code is stored in the computer-readable storage medium, the at least one piece of program code being loaded and executed by a processor to enable a computer to implement any of the point cloud data denoising methods described above.

[0039] On the other hand, a computer program or computer program product is also provided, wherein the computer program or computer program product stores at least one computer instruction, which is loaded and executed by a processor to enable the computer to implement any of the point cloud data denoising methods described above.

[0040] The technical solution provided in this application has at least the following beneficial effects:

[0041] The technical solution provided in this application determines whether there are noise points among multiple feature points by using the laser emission time, echo reception time, and pulse width information of each feature point. If noise points exist among multiple feature points, the feature information of the noise points is removed from the initial point cloud data to obtain target point cloud data, resulting in high accuracy of the obtained target point cloud data. Since the target point cloud data is used to determine the shape of the target object, the accuracy of the determined target object shape is high. The shape of the target object is used to guide the driving of autonomous vehicles, thereby improving the driving safety of autonomous vehicles. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the implementation environment of a point cloud data denoising method provided in an embodiment of this application;

[0044] Figure 2 This is a flowchart of a point cloud data denoising method provided in an embodiment of this application;

[0045] Figure 3 This is a flowchart of a point cloud data denoising method provided in an embodiment of this application;

[0046] Figure 4 This is a schematic diagram of the structure of a point cloud data denoising device provided in an embodiment of this application;

[0047] Figure 5 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application;

[0048] Figure 6 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0050] It should be noted that the terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0051] Figure 1 This is a schematic diagram illustrating the implementation environment of a point cloud data denoising method provided in an embodiment of this application, such as... Figure 1 As shown, the implementation environment includes: terminal device 101, server 102 and lidar 103.

[0052] Among them, the lidar 103 is used to scan the target object to obtain the initial point cloud data. The terminal device 101, the server 102 and the lidar 103 can all be used to denoise the initial point cloud data to obtain the target point cloud data.

[0053] Optionally, after the lidar 103 acquires the initial point cloud data, it sends the initial point cloud data to the terminal device 101. The terminal device 101 then sends the initial point cloud data to the server 102. The server 102 performs noise reduction based on the received initial point cloud data to obtain the target point cloud data.

[0054] Optionally, after acquiring the initial point cloud data, the lidar 103 sends the initial point cloud data to the terminal device 101, and the terminal device 101 performs noise reduction on the received initial point cloud data to obtain the target point cloud data.

[0055] Optionally, after acquiring the initial point cloud data, the lidar 103 performs noise reduction on the initial point cloud data to obtain the target point cloud data.

[0056] Optionally, after acquiring the initial point cloud data, the lidar 103 sends the initial point cloud data to the server 102. The server 102 denoises the received initial point cloud data to obtain the target point cloud data.

[0057] Optionally, after the lidar 103 acquires the initial point cloud data, it sends the initial point cloud data to the server 102. The server 102 then sends the initial point cloud data to the terminal device 101. The terminal device 101 denoises the received initial point cloud data to obtain the target point cloud data.

[0058] The terminal device 101 can be any electronic product capable of human-computer interaction with the user through one or more methods such as a keyboard, touchpad, touchscreen, remote control, voice interaction, or handwriting device. Examples include PCs (Personal Computers), mobile phones, smartphones, PDAs (Personal Digital Assistants), wearable devices, PPCs (Pocket PCs), tablets, smart car systems, smart TVs, and smart speakers. The server 102 can be a single server, a server cluster consisting of multiple servers, or any of the following: a cloud computing platform or a virtualization center. This embodiment of the application does not limit the specific server type. The terminal device 101 and the server 102 communicate via a wired or wireless network, and the lidar 103 communicates with both the terminal device 101 and the server 102 via a wired or wireless network.

[0059] Those skilled in the art should understand that the terminal device 101 and server 102 described above are merely illustrative examples. Other existing or future terminal devices or servers that are applicable to this application should also be included within the scope of protection of this application, and are hereby incorporated by reference.

[0060] This application provides a point cloud data denoising method, which can be applied to the above-mentioned... Figure 1 The implementation environment shown is as follows: Figure 2 The flowchart shown in this embodiment of the present application illustrates a point cloud data denoising method. This method can be implemented by... Figure 1 The terminal device 101 in the middle can be executed, or it can be executed by Figure 1 The server 102 in the middle can be executed, or it can be executed by Figure 1 The lidar 103 in the middle can be executed, and can also be... Figure 1 The interaction between the terminal device 101 and the lidar 103 can also be achieved by... Figure 1 The interaction between server 102 and lidar 103 is implemented using this method. Figure 1 Taking the LiDAR 103 in the middle as an example, such as Figure 2 As shown, the method includes the following steps 201 to 203.

[0061] In step 201, initial point cloud data is acquired, which includes feature information of multiple feature points obtained by the lidar emitting laser pulses toward the target object from multiple directions.

[0062] Each direction corresponds to at least one feature point, and the feature information includes laser emission time, echo reception time, and pulse width information. The laser emission time of any feature point is the time when the lidar emits a laser pulse towards that feature point; the echo reception time of any feature point is the time when the lidar receives the echo reflected from that feature point; and the pulse width information of any feature point indicates the pulse width of the echo reflected from that feature point.

[0063] In one possible implementation, the target object refers to an object whose shape is determined by the LiDAR through scanning. The target object can be any size and any type of object that requires measurement; this application embodiment does not limit this. For example, the target object may be a hill, a table or chair, a wall, a vehicle, etc.

[0064] A lidar system includes a laser transmitter and a laser receiver. The laser transmitter generates multiple laser pulses, each emitted in a different direction, and directs these pulses towards a target object. When there are no other objects between the lidar and the target object, the laser pulses emitted from the laser transmitter in any direction hit a first reference point on the target object, which is then used as the feature point corresponding to that direction. The first reference point reflects the laser pulses received, and the reflected pulses are sent back to the laser receiver as an echo. The laser receiver receives the echo from the first reference point. The laser transmitter records the time of emitting the laser pulses towards the target object as the laser emission time, and the laser receiver records the time of receiving the echo from the first reference point as the echo reception time. Thus, the laser emission time and echo reception time of the first reference point are obtained. This application does not limit the number of laser pulses emitted by the laser transmitter towards the target object.

[0065] When there is a first object between the lidar and the target object, the first object is an object that the laser pulse cannot penetrate, such as a wall or a person. The laser pulse emitted by the laser emitter towards the target object in any direction will be blocked by the first object, so that the laser pulse hits the second reference point of the first object. The second reference point is used as the feature point corresponding to any direction. The time when the laser emitter emits the laser pulse towards the target object is used as the laser emission time of the second reference point. The time when the laser receiver receives the echo reflected from the second reference point is used as the echo reception time of the second reference point.

[0066] When a second object exists between the lidar and the target object, and this second object is one that the laser pulse can penetrate (e.g., a raindrop), the laser pulse emitted by the laser emitter in any direction towards the target object will hit a third reference point on the second object. The laser pulse will also penetrate the third reference point and hit a fourth reference point on the target object. In this case, the third and fourth reference points are considered as feature points corresponding to any direction. The time when the laser emitter emits the laser pulse is taken as the laser emission time of the third and fourth reference points. The time when the laser receiver receives the echo reflected from the third reference point is taken as the echo reception time of the third reference point, and the time when the laser receiver receives the echo reflected from the fourth reference point is taken as the echo reception time of the fourth reference point.

[0067] This application does not limit the scanning rate of the LiDAR on the target object. The scanning rate can be set based on experience or adjusted based on the implementation environment. Optionally, the LiDAR can also choose any scanning method to scan the target object. The scanning methods of the LiDAR include, but are not limited to, line scanning, conical scanning, or fiber optic scanning. After the LiDAR scans the target object and obtains initial point cloud data, environmental factors, such as rain, snow, or sandstorms, will cause noise data in the initial point cloud data, resulting in low accuracy. Therefore, it is necessary to perform noise reduction processing on the initial point cloud data to obtain more accurate target point cloud data.

[0068] In step 202, noise points are determined among the multiple feature points based on the laser emission time, echo reception time, and pulse width information of the multiple feature points.

[0069] In one possible implementation, the process of determining noise points among multiple feature points based on the laser emission time, echo reception time, and pulse width information of multiple feature points includes: determining the distance information of each feature point based on the laser emission time and echo reception time of each feature point, wherein the distance information of any feature point indicates the distance between any feature point and the lidar; and determining noise points among multiple feature points based on the distance information and pulse width information of each feature point.

[0070] Optionally, the process of determining the distance information of each feature point based on the laser emission time and echo reception time of each feature point includes: determining the laser flight time of any feature point based on the laser emission time and echo reception time of any feature point; and determining the distance information of any feature point based on the laser flight speed and the laser flight time of any feature point.

[0071] The difference between the echo reception time and the laser emission time of any feature point is taken as the laser flight time of that feature point. For example, if the echo reception time of any feature point is 14:28:30 and the laser emission time of any feature point is 14:28:20, then the laser flight time of any feature point is 10 seconds.

[0072] Since the laser flight time at any feature point refers to the time required for the lidar to receive the echo after the lidar emits a laser beam at that feature point and reflects it back to the lidar, and for the lidar to receive the echo reflected from that feature point, the distance information of any feature point can be determined in two ways based on the laser flight speed and the laser flight time at any feature point.

[0073] Method 1: Determine the first distance based on the laser flight speed and the laser flight time of any feature point, and then determine the distance information of any feature point based on the first distance.

[0074] Here, the first distance refers to the sum of the distance from the lidar to any feature point and the distance from any feature point to the lidar. Optionally, the product of the laser flight speed and the laser flight time at any feature point can be used as the first distance. The process of determining the distance information of any feature point based on the first distance includes: taking half of the first distance as the distance information of any feature point.

[0075] For example, the laser flight speed is V, and the laser flight time at any feature point is T. Based on the laser flight speed and the laser flight time at any feature point, the first distance is determined as S = V * T. S / 2 is used as the distance information for any feature point.

[0076] Optionally, the distance information S of any feature point can be determined according to the following formula (1) based on the laser flight speed and the laser flight time of any feature point.

[0077]

[0078] In the above formula (1), V is the laser flight speed and T is the laser flight time at any feature point.

[0079] Method 2: Determine the one-way duration based on the laser flight time of any feature point, and determine the distance information of any feature point based on the laser flight speed and the one-way duration.

[0080] Here, one-way time refers to the time required for the laser to travel from the lidar to any feature point, or the time required for the echo to travel from any feature point to the lidar. Optionally, half of the laser's flight time to any feature point can be used as the one-way time. The process of determining the distance information of any feature point based on the laser's flight speed and one-way time includes: using the product of the laser's flight speed and the one-way time as the distance information of any feature point.

[0081] For example, the laser flight speed is V, the laser flight time at any feature point is T, the one-way time determined based on the laser flight time at any feature point is T / 2, and V*T / 2 is used as the distance information of any feature point.

[0082] Optionally, the distance information S of any feature point can be determined according to the following formula (2) based on the laser flight speed and the laser flight time of any feature point.

[0083]

[0084] In the above formula (2), V is the laser flight speed and T is the laser flight time at any feature point.

[0085] It should be noted that any of the above implementation methods can be chosen to determine the distance information of any feature point, and this application embodiment does not limit this. Since the initial point cloud data includes the laser emission time and echo reception time of multiple feature points, it is necessary to determine the distance information of each feature point based on the laser emission time and echo reception time of each feature point. Therefore, the process of determining the distance information of each feature point can be synchronous, or the distance information of each feature point can be determined sequentially according to the time of echo reception of each feature point. This application embodiment does not limit the order in which the distance information of each feature point is determined.

[0086] It should also be noted that pulse width information is positively correlated with reflectivity and inversely correlated with distance information. That is, the stronger the laser reflectivity of the feature point, the smaller the distance information of the feature point, and the larger the pulse width information of the feature point.

[0087] In one possible implementation, noise points can be determined from multiple feature points in two ways, based on the distance information and pulse width information of each feature point.

[0088] The first method involves determining a reference point among multiple feature points. Based on the existence of candidate points among the reference points whose pulse width information is greater than the pulse width threshold, these candidate points are treated as noise points.

[0089] The distance to the reference point is less than a distance threshold. The distance threshold is set based on experience or adjusted according to the implementation environment; this embodiment does not limit this. For example, the distance threshold is 8 meters. The pulse width threshold is also set based on experience or adjusted according to the implementation environment; this embodiment does not limit this either.

[0090] Because noise points are small targets with limited reflective areas, lidar has limited detection capabilities. Beyond a certain distance threshold, the lidar cannot detect whether a feature point is a noise point. Therefore, a distance threshold needs to be set to detect only feature points within the lidar's range, determining whether noise points exist among them. The pulse width of non-noise points should be less than the pulse width threshold. Since noise points are small, low-reflectivity targets, the energy of the laser reflected from them is much less than that reflected from normal objects, causing the lidar to miss the falling edge of the noise point, resulting in the noise point's pulse width exceeding the threshold.

[0091] In one possible implementation, after treating candidate points as noise points, the reflection energy information of the target point can be determined based on the distance information and pulse width information of the target point. If the reflection energy information of the target point is less than the reflection energy threshold, the target point is treated as a noise point.

[0092] Among them, the target point is the feature point in the reference point excluding the candidate points, and the reflection energy information of the target point indicates the energy reflected by the target point.

[0093] The process of determining the reflected energy information of a target point based on its distance and pulse width information includes: multiplying the distance and pulse width information of the target point as the reflected energy information of the target point.

[0094] For example, the reflected energy information ρ of the target point is determined according to the distance information and pulse width information of the target point, according to the following formula (3).

[0095] ρ=R*PW Formula (3)

[0096] In the above formula (3), R is the distance information of the target point, and PW is the pulse width information of the target point.

[0097] Optionally, before treating the target point as a noise point if the reflected energy information based on the target point is less than the reflected energy threshold, it is also necessary to determine the reflected energy threshold. This application does not limit the process for determining the reflected energy threshold.

[0098] Optionally, the process of determining the reflection energy threshold includes: obtaining a training dataset, which includes feature information of multiple first points and feature information of multiple second points, wherein the multiple first points are noise points and the multiple second points are non-noise points; and determining the reflection energy threshold based on the feature information of the multiple first points and the feature information of the multiple second points.

[0099] Optionally, the process of determining the reflection energy threshold based on the feature information of multiple first points and multiple second points includes: determining the reflection energy information of each first point based on the feature information of multiple first points; determining the reflection energy information of each second point based on the feature information of multiple second points; determining a first threshold; determining the first number of points whose reflection energy information is less than the first threshold based on the reflection energy information of each first point and each second point; determining a first loss value based on the number of first points and the first number; and using the first threshold as the reflection energy threshold based on the first loss value being greater than the loss threshold.

[0100] If the first loss value is not greater than the loss threshold, the first threshold is adjusted to obtain the second threshold; the second number of points whose reflected energy information is less than the second threshold is determined, and the second loss value is determined based on the number of the first points and the second number. If the second loss value is greater than the loss threshold, the second threshold is used as the reflected energy threshold.

[0101] If the second loss value is not greater than the loss threshold, the second threshold is adjusted until the loss value determined by the adjusted threshold is greater than the loss threshold. The adjusted threshold is then used as the reflection energy threshold.

[0102] For example, the process of obtaining the training dataset includes: setting the distance threshold to 8 meters, placing 18% reflectivity plates and 90% reflectivity plates at distances of 5 meters and 10 meters from the lidar, simulating rain and fog by spraying water with a sprinkler at distances of 1 meter, 3 meters, 5 meters, and 7 meters, collecting feature information of the training points, and determining whether each training point is the first point or the second point by manual annotation.

[0103] In one possible implementation, the process of determining the reflection energy information of each first point based on the feature information of multiple first points includes: for any one of the multiple first points, determining the distance information of any one first point based on the laser emission time and echo reception time of any one first point; and determining the reflection energy information of any one first point based on the distance information and pulse width information of any one first point.

[0104] The process of determining the distance information of any first point based on the laser emission time and echo reception time of any first point is similar to the process of determining the distance information of each feature point based on the laser emission time and echo reception time of each feature point described above. Similarly, the process of determining the reflection energy information of any first point based on its distance information and pulse width information is similar to the process of determining the reflection energy information of the target point based on its distance information and pulse width information described above. These details will not be described again in this embodiment. The process of determining the reflection energy information of each second point based on the feature information of multiple second points is similar to the process of determining the reflection energy information of each first point based on the feature information of multiple first points, and will not be elaborated upon here either.

[0105] The process of determining the first loss value based on the number of first points and the first number includes taking the quotient of the first number and the number of first points as the first loss value. The loss threshold is set based on experience or adjusted according to the implementation environment; this embodiment does not limit this. For example, the loss threshold is 90.

[0106] For example, the initial point cloud data includes 10 feature points. Based on the laser emission time and echo reception time of each feature point, the distance information of each feature point is determined. The distance information and pulse width information of each feature point are shown in Table 1 below, and will not be elaborated further here. The distance threshold is 8 meters, and the pulse width threshold is 60. Based on the distance thresholds of each feature point, feature points 4, 5, 6, 7, and 8 are determined as reference points from the 10 feature points. Since there are candidate points among the reference points with pulse width information greater than the pulse width threshold, feature points 4, 6, 7, and 8 are considered noise points.

[0107] Table 1

[0108]

[0109]

[0110] Optionally, after designating feature points 4, 6, 7, and 8 as noise points, the reflection energy information of feature point 5 can be determined based on its distance and pulse width information. The reflection energy information of feature point 5 is 6 * 60 = 360. The reflection energy threshold is 400. Since the reflection energy information of feature point 5 is less than the reflection energy threshold, feature point 5 is also considered a noise point. If the reflection energy threshold is 300, since the reflection energy information of feature point 5 is greater than the reflection energy threshold, it is unnecessary to consider feature point 5 as a noise point.

[0111] The second method involves determining reference points among multiple feature points, and then determining the reflection energy information of each reference point based on its distance and pulse width information. Reference points whose reflection energy information is less than the reflection energy threshold are designated as noise points.

[0112] In this context, the distance information of the reference point is less than the distance threshold. The reflected energy information of any reference point indicates the energy reflected by that reference point. Before classifying reference points whose reflected energy information is less than the reflected energy threshold as noise points, it is necessary to determine the reflected energy threshold. The process of determining the reflected energy threshold has been described in the first method and will not be repeated here.

[0113] The process of determining the reflected energy information of each reference point based on the distance information and pulse width information of each reference point includes: for any reference point among multiple reference points, the product of the distance information and the pulse width information of any reference point is taken as the reflected energy information of any reference point.

[0114] It should be noted that any of the above methods can be used to determine noise points among multiple feature points, and this application embodiment does not limit this. Compared with the second method, the first method determines noise points more comprehensively.

[0115] In step 203, the feature information of the noise points is removed from the initial point cloud data to obtain the target point cloud data.

[0116] Target point cloud data is used to determine the shape of the target object. After determining the shape of the target object, the autonomous vehicle is then controlled to drive based on the shape of the target object. If the shape of the target object indicates that the target object is an obstacle, the autonomous vehicle is controlled to bypass the target object; if the shape of the target object indicates that the target object is not an obstacle, the autonomous vehicle is controlled to pass through the target object.

[0117] For example, the initial point cloud data includes 10 feature points, namely feature point 1 to feature point 10. Among them, feature point 4, feature point 6, feature point 7 and feature point 8 are noise points. Therefore, the feature information of feature point 4, feature point 6, feature point 7 and feature point 8 is removed from the initial point cloud data to obtain the target point cloud data.

[0118] The above method determines the presence of noise points among multiple feature points by using the laser emission time, echo reception time, and pulse width information of each feature point. If noise points are found, their feature information is removed from the initial point cloud data to obtain the target point cloud data, resulting in high accuracy. Since the target point cloud data is used to determine the shape of the target object, the accuracy of the determined target object shape is high. The shape of the target object guides the driving of autonomous vehicles, thereby improving the driving safety of autonomous vehicles.

[0119] In one possible implementation, Figure 3 This is a flowchart of another point cloud data denoising method provided in this application embodiment. This method is implemented through interaction between an electronic device and a LiDAR. The electronic device can be a terminal device or a server, such as... Figure 3 As shown, the method includes the following steps 301 to 305.

[0120] In step 301, the lidar acquires initial point cloud data.

[0121] The initial point cloud data includes feature information of multiple feature points obtained by the lidar emitting laser pulses towards the target object in multiple directions. Each direction corresponds to at least one feature point. The feature information includes laser emission time, echo reception time, and pulse width information. The laser emission time of any feature point is the time when the lidar emits a laser pulse towards that feature point. The echo reception time of any feature point is the time when the lidar receives the echo reflected from that feature point. The pulse width information of any feature point indicates the pulse width of the echo reflected from that feature point.

[0122] In one possible implementation, the process of the lidar acquiring the initial point cloud data of the scanned target object has been described in step 201 above, and will not be repeated here.

[0123] In step 302, the lidar sends initial point cloud data to the electronic device.

[0124] The lidar and electronic equipment communicate via a wired or wireless network. The lidar is installed on the autonomous vehicle, and the electronic equipment is either installed on the autonomous vehicle or capable of remotely controlling the autonomous vehicle; this application does not limit this. After acquiring initial point cloud data, the lidar sends the initial point cloud data to the electronic equipment so that the electronic equipment can acquire the initial point cloud data.

[0125] In step 303, the electronic device receives the initial point cloud data sent by the lidar and determines the distance information of each feature point based on the laser emission time and echo reception time of multiple feature points.

[0126] In one possible implementation, the distance information of any feature point indicates the distance between that feature point and the lidar. The process of determining the distance information of each feature point based on the laser emission time and echo reception time of multiple feature points has been described in step 202 above and will not be repeated here.

[0127] In step 304, the electronic device determines noise points among multiple feature points based on the distance information and pulse width information of each feature point.

[0128] In one possible implementation, the process of determining noise points among multiple feature points based on the distance information and pulse width information of each feature point has been described in step 202 above and will not be repeated here.

[0129] In step 305, the electronic device removes the feature information of the noise points from the initial point cloud data to obtain the target point cloud data.

[0130] In one possible implementation, the process of removing the feature information of noise points from the initial point cloud data to obtain the target point cloud data has been described in step 203 above and will not be repeated here.

[0131] Since electronic devices can control autonomous vehicles, after acquiring target point cloud data, they can also determine the shape of the target object based on this data and control the autonomous vehicle's movement accordingly. If the target object's shape indicates it is an obstacle, the electronic device sends a first control command to the autonomous vehicle, instructing it to bypass the target object. If the target object's shape indicates it is not an obstacle, the electronic device sends a second control command to the autonomous vehicle, instructing it to pass the target object.

[0132] Figure 4 The diagram shown is a structural schematic of a point cloud data denoising device provided in an embodiment of this application. Figure 4 As shown, the device includes:

[0133] The acquisition module 401 is used to acquire initial point cloud data. The initial point cloud data includes feature information of multiple feature points obtained by the lidar emitting laser pulses towards the target object in multiple directions. Each direction corresponds to at least one feature point. The feature information includes laser emission time, echo reception time and pulse width information. The laser emission time of any feature point is the time when the lidar emits a laser pulse towards any feature point. The echo reception time of any feature point is the time when the lidar receives the echo reflected by any feature point. The pulse width information of any feature point indicates the pulse width of the echo reflected by any feature point.

[0134] The determination module 402 is used to determine noise points among multiple feature points based on the laser emission time, echo reception time, and pulse width information of multiple feature points;

[0135] The noise removal module 403 is used to remove the feature information of noise points from the initial point cloud data to obtain target point cloud data, which is used to determine the shape of the target object.

[0136] In one possible implementation, the determining module 402 is used to determine the distance information of each feature point based on the laser emission time and echo reception time of each feature point, wherein the distance information of any feature point indicates the distance between any feature point and the lidar; and to determine noise points among multiple feature points based on the distance information and pulse width information of each feature point.

[0137] In one possible implementation, the determining module 402 is used to determine a reference point among multiple feature points, wherein the distance information of the reference point is less than a distance threshold; and based on the existence of candidate points among the reference points whose pulse width information is greater than the pulse width threshold, the candidate points are treated as noise points.

[0138] In one possible implementation, the determining module 402 is further configured to determine the reflected energy information of the target point based on the distance information and pulse width information of the target point. The reflected energy information of the target point indicates the energy reflected by the target point, and the target point is a feature point in the reference points other than the candidate points. Based on the fact that the reflected energy information of the target point is less than the reflected energy threshold, the target point is regarded as a noise point.

[0139] In one possible implementation, the determining module 402 is used to take the product of the distance information and the pulse width information of the target point as the reflected energy information of the target point.

[0140] In one possible implementation, the determining module 402 is used to determine a reference point among multiple feature points, wherein the distance information of the reference point is less than a distance threshold; determine the reflection energy information of each reference point based on the distance information and pulse width information of each reference point, wherein the reflection energy information of any reference point indicates the energy reflected by any reference point; and designate reference points whose reflection energy information is less than the reflection energy threshold as noise points.

[0141] In one possible implementation, the determining module 402 is used to determine the laser flight time of any feature point among a plurality of feature points based on the laser emission time and echo reception time of any feature point; and to determine the distance information of any feature point based on the laser flight speed and the laser flight time of any feature point.

[0142] The aforementioned device determines the presence of noise points among multiple feature points by analyzing the laser emission time, echo reception time, and pulse width information of each feature point. If noise points are found, their feature information is removed from the initial point cloud data to obtain the target point cloud data, resulting in high accuracy. Since the target point cloud data is used to determine the shape of the target object, the accuracy of the determined shape is high. The shape of the target object guides the driving of autonomous vehicles, thereby improving their driving safety.

[0143] It should be understood that the above-described apparatus is only illustrated by the division of the functional modules described above when implementing its functions. In practical applications, the functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0144] Figure 5 This illustration shows a structural block diagram of a terminal device 500 provided in an exemplary embodiment of this application. The terminal device 500 may be a portable mobile terminal, such as a smartphone, tablet computer, MP3 player (Moving Picture Experts Group Audio Layer III), MP4 player (Moving Picture Experts Group Audio Layer IV), laptop computer, or desktop computer. The terminal device 500 may also be referred to as a user device, portable terminal, laptop terminal, desktop terminal, or other names.

[0145] Typically, terminal device 500 includes a processor 501 and a memory 502.

[0146] Processor 501 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 501 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 501 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 501 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 501 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0147] The memory 502 may include one or more computer-readable storage media, which may be non-transitory. The memory 502 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 502 are used to store at least one instruction, which is executed by the processor 501 to implement the point cloud data denoising method provided in the method embodiments of this application.

[0148] In some embodiments, the terminal device 500 may also optionally include a peripheral device interface 503 and at least one peripheral device. The processor 501, memory 502, and peripheral device interface 503 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 503 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: a radio frequency circuit 504, a display screen 505, a camera assembly 506, an audio circuit 507, a positioning assembly 508, and a power supply 509.

[0149] Peripheral device interface 503 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 501 and memory 502. In some embodiments, processor 501, memory 502 and peripheral device interface 503 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 501, memory 502 and peripheral device interface 503 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0150] The radio frequency (RF) circuit 504 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 504 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 504 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 504 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 504 can communicate with other terminal devices through at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit 504 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0151] Display screen 505 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 505 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 501 for processing. In this case, display screen 505 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 505, disposed on the front panel of terminal device 500; in other embodiments, there may be at least two display screens, disposed on different surfaces of terminal device 500 or in a folded design; in other embodiments, display screen 505 may be a flexible display screen, disposed on a curved or folded surface of terminal device 500. Furthermore, display screen 505 may be configured as a non-rectangular irregular shape, i.e., a non-rectangular screen. Display screen 505 may be made of materials such as LCD (Liquid Crystal Display) or OLED (Organic Light-Emitting Diode).

[0152] The camera assembly 506 is used to acquire images or videos. Optionally, the camera assembly 506 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the terminal device 500, and the rear-facing camera is located on the back of the terminal device 500. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 506 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cool light flash, which can be used for light compensation at different color temperatures.

[0153] The audio circuit 507 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting the sound waves into electrical signals that are input to the processor 501 for processing, or input to the radio frequency circuit 504 to achieve voice communication. For stereo sound acquisition or noise reduction purposes, there may be multiple microphones, each located at a different part of the terminal device 500. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert the electrical signals from the processor 501 or the radio frequency circuit 504 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 507 may also include a headphone jack.

[0154] The positioning component 508 is used to locate the current geographical location of the terminal device 500 in order to enable navigation or LBS (Location Based Service). The positioning component 508 can be a positioning component based on the US GPS (Global Positioning System), China's BeiDou system, Russia's Granas system, or the European Union's Galileo system.

[0155] Power supply 509 is used to supply power to the various components in terminal device 500. Power supply 509 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 509 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0156] In some embodiments, the terminal device 500 further includes one or more sensors 510. The one or more sensors 510 include, but are not limited to: an accelerometer 511, a gyroscope 512, a pressure sensor 513, a fingerprint sensor 514, an optical sensor 515, and a proximity sensor 516.

[0157] Accelerometer 511 can detect the magnitude of acceleration along the three coordinate axes of a coordinate system established by terminal device 500. For example, accelerometer 511 can be used to detect the components of gravitational acceleration along the three coordinate axes. Processor 501 can control display screen 505 to display the user interface in either a landscape or portrait view based on the gravitational acceleration signal acquired by accelerometer 511. Accelerometer 511 can also be used for games or for acquiring user motion data.

[0158] The gyroscope sensor 512 can detect the orientation and rotation angle of the terminal device 500. The gyroscope sensor 512, in conjunction with the accelerometer sensor 511, can collect 3D motion data from the user on the terminal device 500. Based on the data collected by the gyroscope sensor 512, the processor 501 can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0159] The pressure sensor 513 can be disposed on the side bezel of the terminal device 500 and / or on the lower layer of the display screen 505. When the pressure sensor 513 is disposed on the side bezel of the terminal device 500, it can detect the user's grip signal on the terminal device 500, and the processor 501 can perform left / right hand recognition or quick operation based on the grip signal collected by the pressure sensor 513. When the pressure sensor 513 is disposed on the lower layer of the display screen 505, the processor 501 can control the operable controls on the UI interface based on the user's pressure operation on the display screen 505. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0160] The fingerprint sensor 514 is used to collect the user's fingerprint. The processor 501 identifies the user's identity based on the fingerprint collected by the fingerprint sensor 514, or the fingerprint sensor 514 identifies the user's identity based on the collected fingerprint. When the user's identity is identified as trusted, the processor 501 authorizes the user to perform relevant sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, making payments, and changing settings. The fingerprint sensor 514 can be located on the front, back, or side of the terminal device 500. When the terminal device 500 has a physical button or manufacturer logo, the fingerprint sensor 514 can be integrated with the physical button or manufacturer logo.

[0161] An optical sensor 515 is used to collect ambient light intensity. In one embodiment, the processor 501 can control the display brightness of the display screen 505 based on the ambient light intensity collected by the optical sensor 515. Specifically, when the ambient light intensity is high, the display brightness of the display screen 505 is increased; when the ambient light intensity is low, the display brightness of the display screen 505 is decreased. In another embodiment, the processor 501 can also dynamically adjust the shooting parameters of the camera assembly 506 based on the ambient light intensity collected by the optical sensor 515.

[0162] The proximity sensor 516, also known as a distance sensor, is typically located on the front panel of the terminal device 500. The proximity sensor 516 is used to detect the distance between the user and the front of the terminal device 500. In one embodiment, when the proximity sensor 516 detects that the distance between the user and the front of the terminal device 500 is gradually decreasing, the processor 501 controls the display screen 505 to switch from a screen-on state to a screen-off state; when the proximity sensor 516 detects that the distance between the user and the front of the terminal device 500 is gradually increasing, the processor 501 controls the display screen 505 to switch from a screen-off state to a screen-on state.

[0163] Those skilled in the art will understand that Figure 5 The structure shown does not constitute a limitation on the terminal device 500, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0164] Figure 6 This is a schematic diagram of the server structure provided in the embodiments of this application. The server 600 can vary considerably due to different configurations or performance. It may include one or more Central Processing Units (CPUs) 601 and one or more memories 602. The one or more memories 602 store at least one line of program code, which is loaded and executed by the one or more processors 601 to implement the point cloud data denoising methods provided in the various method embodiments described above. Of course, the server 600 may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server 600 may also include other components for implementing device functions, which will not be elaborated here.

[0165] In an exemplary embodiment, a computer-readable storage medium is also provided, which stores at least one piece of program code that is loaded and executed by a processor to enable a computer to implement any of the point cloud data denoising methods described above.

[0166] Optionally, the aforementioned computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0167] In an exemplary embodiment, a computer program or computer program product is also provided, which stores at least one computer instruction, which is loaded and executed by a processor to enable the computer to implement any of the point cloud data denoising methods described above.

[0168] It should be noted that all information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.), and signals involved in this application have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the initial point cloud data involved in this application were all obtained with full authorization.

[0169] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0170] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0171] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for point cloud data denoising, characterized in that, The method comprises: acquiring initial point cloud data, the initial point cloud data comprising feature information of a plurality of feature points obtained by a laser radar emitting laser pulses in a plurality of directions to a target object, any direction corresponding to at least one feature point, the feature information comprising laser emission time, echo reception time and pulse width information, the laser emission time of any feature point being the time at which the laser radar emits a laser pulse to the any feature point, the echo reception time of the any feature point being the time at which the laser radar receives an echo reflected by the any feature point, and the pulse width information of the any feature point indicating the pulse width of the echo reflected by the any feature point; determining distance information of each feature point according to the laser emission time and the echo reception time of each feature point, the distance information of any feature point indicating the distance between the any feature point and the laser radar; determining a reference point in the plurality of feature points, the distance information of the reference point being less than a distance threshold value; based on a candidate point in the reference point having pulse width information greater than a pulse width threshold value, regarding the candidate point as a noise point; determining reflection energy information of a target point according to the distance information and the pulse width information of the target point, the reflection energy information of the target point indicating the energy reflected by the target point, the target point being a feature point in the reference point other than the candidate point; based on the reflection energy information of the target point being less than a reflection energy threshold value, regarding the target point as the noise point; or, determining reflection energy information of each reference point according to the distance information and the pulse width information of each reference point, the reflection energy information of any reference point indicating the energy reflected by the any reference point; regarding a reference point in the reference points having reflection energy information less than a reflection energy threshold value as a noise point; removing the feature information of the noise point from the initial point cloud data to obtain target point cloud data, the target point cloud data being used to determine the morphology of the target object.

2. The method of claim 1, wherein, The determining of the reflection energy information of the target point according to the distance information and the pulse width information of the target point comprises: multiplying the distance information and the pulse width information of the target point to obtain the reflection energy information of the target point.

3. The method according to claim 1 or 2, characterized in that, The determining of the distance information of each feature point according to the laser emission time and the echo reception time of each feature point comprises: for any feature point in the plurality of feature points, determining the laser flight time of the any feature point according to the laser emission time and the echo reception time of the any feature point; determining the distance information of the any feature point according to the laser flight speed and the laser flight time of the any feature point.

4. A point cloud data denoising apparatus, characterized by comprising: The device comprises: An acquisition module is configured to acquire initial point cloud data, the initial point cloud data comprising feature information of a plurality of feature points obtained by a laser radar emitting laser pulses in a plurality of directions to a target object, any direction corresponding to at least one feature point, the feature information comprising laser emission time, echo reception time and pulse width information, the laser emission time of any feature point being the time when the laser radar emits a laser pulse to the any feature point, the echo reception time of the any feature point being the time when the laser radar receives an echo reflected by the any feature point, and the pulse width information of the any feature point indicating the pulse width of the echo reflected by the any feature point. A determination module is configured to determine distance information of each feature point according to the laser emission time and the echo reception time of each feature point, the distance information of any feature point indicating the distance between the any feature point and the laser radar; determine a reference point in the plurality of feature points, the distance information of the reference point being less than a distance threshold; take a candidate point in the reference point as a noise point based on the pulse width information of the candidate point being greater than a pulse width threshold; determine reflection energy information of a target point according to the distance information and the pulse width information of the target point, the reflection energy information of the target point indicating the energy reflected by the target point, the target point being a feature point in the reference point other than the candidate point; take the target point as the noise point based on the reflection energy information of the target point being less than a reflection energy threshold; or determine reflection energy information of each reference point according to the distance information and the pulse width information of each reference point, the reflection energy information of any reference point indicating the energy reflected by the any reference point; take a reference point in the reference points as the noise point based on the reflection energy information of the reference point being less than the reflection energy threshold. A removal module is configured to remove the feature information of the noise point from the initial point cloud data to obtain target point cloud data, the target point cloud data being used to determine the morphology of the target object.

5. A computer device, comprising: The computer device comprises a processor and a memory, the memory storing at least one program code, the at least one program code being loaded and executed by the processor to enable the computer device to implement the point cloud data denoising method according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium stores at least one program code, the at least one program code being loaded and executed by the processor to enable the computer to implement the point cloud data denoising method according to any one of claims 1 to 3.

Images