Method and device for filtering out outliers of laser point cloud, terminal equipment and storage medium

Abstract

The application is suitable for the field of radar technology, and provides a method and device for filtering out a trailing point of a laser point cloud, a terminal device and a storage medium. The method comprises the following steps: acquiring laser point cloud data to be processed collected by a laser radar; determining at least one adjacent point and at least one reference window of a measured point in a horizontal direction; determining a position feature between the measured point and each adjacent point and each reference window; determining whether the measured point is a trailing point according to the position feature between the measured point and each adjacent point and each reference window; and removing the measured point from the laser point cloud data in the case that the measured point is a trailing point. Thus, the trailing point in the laser point cloud is effectively identified and filtered out, so that the influence of the trailing point on the identification accuracy of a point cloud backend algorithm is eliminated, and the accuracy of a ranging, target identification and other backend algorithm is improved.

Description

Technical Field

[0001] This application belongs to the field of radar technology, and in particular relates to a method, apparatus, terminal equipment and computer-readable storage medium for filtering out drag points in laser point clouds. Background Technology

[0002] As a new type of sensing device, lidar accurately measures the three-dimensional coordinates and reflectivity information of targets within the detection area by emitting multiple laser beams into space, thereby effectively reconstructing the measured area.

[0003] In related technologies, because the emitted beam of a lidar has a certain divergence angle, when a beam hits the edge of object A, some of the light will hit behind object A. If there is another object B at a close distance behind object A, the echo signals from these two objects will be superimposed, leading to inaccurate timing and pulse width information. This manifests as a point cloud existing between the two objects; these points are commonly referred to as drag points. The presence of drag points affects the accuracy of subsequent target recognition algorithms or ranging in the point cloud. Therefore, effectively identifying and filtering drag points in lidar point clouds is a problem that urgently needs to be solved. Summary of the Invention

[0004] This application provides a method, apparatus, terminal device, and storage medium for filtering drag points in laser point clouds, which can effectively identify and filter drag points in laser point clouds, thereby improving the accuracy of subsequent ranging or target identification using laser point clouds.

[0005] In a first aspect, embodiments of this application provide a method for filtering dragged points in laser point clouds, comprising: acquiring laser point cloud data to be processed collected by a lidar, wherein the laser point cloud data contains multiple point cloud points; determining at least one adjacent point and at least one reference window of the measured point in the horizontal direction, wherein the measured point is any point cloud point in the laser point cloud data, and the reference window contains multiple reference points; determining the positional characteristics between the measured point and each adjacent point and each reference window respectively; determining whether the measured point is a dragged point based on the positional characteristics between the measured point and each adjacent point and each reference window respectively; and removing the measured point from the laser point cloud data if the measured point is a dragged point.

[0006] In one possible implementation of the first aspect, the aforementioned positional features include distance features and angle features; correspondingly, determining the positional features between the measured point and each adjacent point and each reference window includes:

[0007] Determine the distance characteristics between the measured point and each adjacent point, and the distance characteristics corresponding to each reference window;

[0008] Determine the angular characteristics between the measured point and each adjacent point and each reference window.

[0009] Optionally, in another possible implementation of the first aspect, determining the distance features between the measured point and each adjacent point, and the distance features corresponding to each reference window, includes:

[0010] Determine the distance between the measured point and each adjacent point;

[0011] Determine the distance range corresponding to each reference point in each reference window;

[0012] Accordingly, the above determination of the angular features between the measured point and each adjacent point and each reference window includes:

[0013] Determine the coplanar angle characteristics between the measured point and each reference window, where the coplanar angle characteristics refer to the angle between the vector between the reference window and the measured point and the vector corresponding to the reference window;

[0014] Determine the collinear angle features between the measured point and each adjacent point. The collinear angle features refer to the angle between the vector corresponding to the measured point and the origin and the vector corresponding to the measured point and the adjacent point. The origin refers to the position of the lidar in the lidar point cloud data.

[0015] Optionally, in another possible implementation of the first aspect, the reference window includes a first reference window and a second reference window; correspondingly, determining the distance features between the measured point and each adjacent point, and the distance features corresponding to each reference window, further includes:

[0016] Determine the mean first distance for each reference point in the first reference window;

[0017] Determine the mean of the second distance for each reference point in the second reference window.

[0018] Optionally, in another possible implementation of the first aspect, determining whether a measured point is a drag point based on the positional characteristics of the measured point with each adjacent point and each reference window includes:

[0019] If the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the distance range of each reference point in each reference window is less than the second distance threshold, and each coplanar angle feature does not meet the coplanar condition, then the measured point is determined to be a drag point. The coplanar condition is used to determine whether the measured point is coplanar with any reference window.

[0020] Otherwise, the measured point is determined to be a normal point;

[0021] or,

[0022] If the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the distance range of each reference point in each reference window is less than the second distance threshold, at least one coplanar angle feature satisfies the coplanar condition, and at least one collinear angle feature satisfies the collinear condition, then the measured point is determined to be a drag point. The collinear condition is used to determine whether the origin, the measured point and any adjacent point are collinear.

[0023] Otherwise, the measured point is determined to be a normal point.

[0024] Optionally, in another possible implementation of the first aspect, determining whether a measured point is a drag point based on the positional characteristics of the measured point with each adjacent point and each reference window includes:

[0025] If the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the range of distances corresponding to each reference point in each reference window is less than the second distance threshold, the distance corresponding to the measured point is between the first average distance and the second average distance, and each coplanar angle feature does not meet the coplanar condition, then the measured point is determined to be a drag point. The coplanar condition is used to determine whether the measured point is coplanar with any reference window.

[0026] Otherwise, the measured point is determined to be a normal point;

[0027] or,

[0028] If the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the range of distances corresponding to each reference point in each reference window is less than the second distance threshold, the distance corresponding to the measured point is between the first distance mean and the second distance mean, at least one coplanar angle feature satisfies the coplanar condition, and at least one collinear angle feature satisfies the collinear condition, then the measured point is determined to be a drag point. The collinear condition is used to determine whether the origin, the measured point and any adjacent point are collinear.

[0029] Otherwise, the measured point is determined to be a normal point.

[0030] Optionally, in another possible implementation of the first aspect, it is determined whether the distance corresponding to the measured point is between the first average distance and the second average distance by the following method:

[0031] The fourth distance threshold is determined based on the smaller of the first and second distance averages and the third distance threshold.

[0032] The fifth distance threshold is determined based on the larger of the first and second distance averages and the third distance threshold.

[0033] If the distance to the measured point is greater than the fourth distance threshold and less than the fifth distance threshold, it is determined that the distance to the measured point is between the first average distance and the second average distance.

[0034] If the distance to the measured point is less than or equal to the fourth distance threshold, or the distance to the measured point is greater than or equal to the fifth distance threshold, it is determined that the distance to the measured point is not between the first distance mean and the second distance mean.

[0035] Optionally, in another possible implementation of the first aspect, the aforementioned coplanar angle features include a first coplanar angle feature corresponding to the first reference window and a second coplanar angle feature corresponding to the second reference window; correspondingly, determining the coplanar angle features between the measured point and each reference window includes:

[0036] Determine the first vector corresponding to the first reference window, and the second vector between the measured point and the first reference window;

[0037] The first coplanar angle feature is determined based on the angle between the first vector and the second vector;

[0038] Determine the third vector corresponding to the second reference window, and the fourth vector between the measured point and the second reference window;

[0039] The second coplanar angle feature is determined based on the angle between the third and fourth vectors.

[0040] Optionally, in another possible implementation of the first aspect, whether the first coplanar angle feature and the second coplanar angle feature satisfy the coplanar condition is determined by the following method:

[0041] Determine whether the first coplanar angle feature and the second coplanar angle feature are less than the first angle threshold or greater than the second angle threshold, wherein the second angle threshold is greater than the first angle threshold;

[0042] If the first coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the first coplanar angle feature satisfies the coplanar condition.

[0043] If the first coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the first coplanar angle feature does not meet the coplanar condition.

[0044] If the second coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the second coplanar angle feature satisfies the coplanar condition.

[0045] If the second coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the second coplanar angle feature does not satisfy the coplanar condition.

[0046] Optionally, in another possible implementation of the first aspect, the aforementioned collinear angle feature includes a first collinear angle feature corresponding to the first reference window and a second collinear angle feature corresponding to the second reference window; correspondingly, determining the collinear angle feature between the measured point and each adjacent point includes:

[0047] Determine the fifth vector between the measured point and the origin;

[0048] Determine the sixth vector between the measured point and the first adjacent point, and the seventh vector between the measured point and the second adjacent point;

[0049] The first collinear angle characteristic is determined based on the angle between the fifth and sixth vectors;

[0050] The second collinearity angle characteristic is determined based on the angle between the fifth and seventh vectors.

[0051] Optionally, in another possible implementation of the first aspect, whether the first collinear angle feature and the second collinear angle feature satisfy the collinearity condition is determined by the following method:

[0052] Determine whether the first collinear angle feature and the second collinear angle feature are less than the third angle threshold or greater than the fourth angle threshold;

[0053] If the first collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the first collinear angle feature satisfies the collinearity condition.

[0054] If the first collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the first collinear angle feature does not meet the collinearity condition.

[0055] If the second collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the second collinear angle feature satisfies the collinearity condition.

[0056] If the second collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the second collinear angle feature does not meet the collinearity condition.

[0057] Optionally, in another possible implementation of the first aspect, the reference window does not include adjacent points.

[0058] Secondly, embodiments of this application provide a device for filtering dragged points in laser point clouds, comprising: a first acquisition module for acquiring laser point cloud data to be processed collected by a lidar, wherein the laser point cloud data contains multiple point cloud points; a first determination module for determining at least one adjacent point and at least one reference window of the measured point in the horizontal direction, wherein the measured point is any point cloud point in the laser point cloud data, and the reference window contains multiple reference points; a second determination module for determining the positional characteristics between the measured point and each adjacent point and each reference window respectively; a third determination module for determining whether the measured point is a dragged point based on the positional characteristics between the measured point and each adjacent point and each reference window respectively; and a first removal module for removing the measured point from the laser point cloud data if the measured point is a dragged point.

[0059] In one possible implementation of the second aspect, the aforementioned positional features include distance features and angle features; correspondingly, the aforementioned second determining module includes:

[0060] The first determining unit is used to determine the distance features between the measured point and each adjacent point, and the distance features corresponding to each reference window;

[0061] The second determining unit is used to determine the angular characteristics between the measured point and each adjacent point and each reference window.

[0062] Optionally, in another possible implementation of the second aspect, the aforementioned first determining unit is specifically used for:

[0063] Determine the distance between the measured point and each adjacent point;

[0064] Determine the distance range corresponding to each reference point in each reference window;

[0065] Accordingly, the second determining unit mentioned above is specifically used for:

[0066] Determine the coplanar angle characteristics between the measured point and each reference window, where the coplanar angle characteristics refer to the angle between the vector between the reference window and the measured point and the vector corresponding to the reference window;

[0067] Determine the collinear angle features between the measured point and each adjacent point. The collinear angle features refer to the angle between the vector corresponding to the measured point and the origin and the vector corresponding to the measured point and the adjacent point. The origin refers to the position of the lidar in the lidar point cloud data.

[0068] Optionally, in another possible implementation of the second aspect, the reference window includes a first reference window and a second reference window; correspondingly, the first determining unit is further configured to:

[0069] Determine the mean first distance for each reference point in the first reference window;

[0070] Determine the mean of the second distance for each reference point in the second reference window.

[0071] Optionally, in another possible implementation of the second aspect, the aforementioned third determining module includes:

[0072] The third determining unit is used to determine the measured point as a drag point when the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the distance range corresponding to each reference point in each reference window is less than the second distance threshold, and each coplanar angle feature does not meet the coplanar condition. The coplanar condition is used to determine whether the measured point is coplanar with any reference window.

[0073] The fourth determining unit is used to determine the measured point as a normal point when any of the above conditions are not met;

[0074] or,

[0075] The fifth determining unit is used to determine the measured point as a drag point when the distance between the measured point and at least one adjacent point is greater than a first distance threshold, the distance range corresponding to each reference point in each reference window is less than a second distance threshold, at least one coplanar angle feature satisfies the coplanar condition, and at least one collinear angle feature satisfies the collinear condition. The collinear condition is used to determine whether the origin, the measured point and any adjacent point are collinear.

[0076] The sixth determining unit is used to determine the measured point as a normal point when none of the above conditions are met.

[0077] Optionally, in another possible implementation of the second aspect, the aforementioned third determining module includes:

[0078] The seventh determining unit is used to determine the measured point as a drag point if the distance between the measured point and at least one adjacent point is greater than a first distance threshold, the distance range corresponding to each reference point in each reference window is less than a second distance threshold, the distance corresponding to the measured point is between the first average distance and the second average distance, and each coplanar angle feature does not meet the coplanar condition. The coplanar condition is used to determine whether the measured point is coplanar with any reference window.

[0079] The eighth determining unit is used to determine the measured point as a normal point when any of the above conditions are not met;

[0080] or,

[0081] The ninth determining unit is used to determine the measured point as a drag point if the distance between the measured point and at least one adjacent point is greater than a first distance threshold, the distance range corresponding to each reference point in each reference window is less than a second distance threshold, the distance corresponding to the measured point is between the first distance mean and the second distance mean, at least one coplanar angle feature satisfies the coplanar condition, and at least one collinear angle feature satisfies the collinear condition. The collinear condition is used to determine whether the origin, the measured point and any adjacent point are collinear.

[0082] The tenth determining unit is used to determine the measured point as a normal point when none of the above conditions are met.

[0083] Optionally, in another possible implementation of the second aspect, the distance corresponding to the measured point is determined as follows:

[0084] The fourth distance threshold is determined based on the smaller of the first and second distance averages and the third distance threshold.

[0085] The fifth distance threshold is determined based on the larger of the first and second distance averages and the third distance threshold.

[0086] If the distance to the measured point is greater than the fourth distance threshold and less than the fifth distance threshold, it is determined that the distance to the measured point is between the first average distance and the second average distance.

[0087] If the distance to the measured point is less than or equal to the fourth distance threshold, or the distance to the measured point is greater than or equal to the fifth distance threshold, it is determined that the distance to the measured point is not between the first distance mean and the second distance mean.

[0088] Optionally, in another possible implementation of the second aspect, the aforementioned coplanar angle feature includes a first coplanar angle feature corresponding to the first reference window and a second coplanar angle feature corresponding to the second reference window; correspondingly, the aforementioned second determining unit is further configured to:

[0089] Determine the first vector corresponding to the first reference window, and the second vector between the measured point and the first reference window;

[0090] The first coplanar angle feature is determined based on the angle between the first vector and the second vector;

[0091] Determine the third vector corresponding to the second reference window, and the fourth vector between the measured point and the second reference window;

[0092] The second coplanar angle feature is determined based on the angle between the third and fourth vectors.

[0093] Optionally, in another possible implementation of the second aspect, whether the first coplanar angle feature and the second coplanar angle feature satisfy the coplanar condition is determined by the following method:

[0094] Determine whether the first coplanar angle feature and the second coplanar angle feature are less than the first angle threshold or greater than the second angle threshold, wherein the second angle threshold is greater than the first angle threshold;

[0095] If the first coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the first coplanar angle feature satisfies the coplanar condition.

[0096] If the first coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the first coplanar angle feature does not meet the coplanar condition.

[0097] If the second coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the second coplanar angle feature satisfies the coplanar condition.

[0098] If the second coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the second coplanar angle feature does not satisfy the coplanar condition.

[0099] Optionally, in another possible implementation of the second aspect, the aforementioned collinear angle feature includes a first collinear angle feature corresponding to the first reference window and a second collinear angle feature corresponding to the second reference window; correspondingly, the aforementioned second determining unit is further configured to:

[0100] Determine the fifth vector between the measured point and the origin;

[0101] Determine the sixth vector between the measured point and the first adjacent point, and the seventh vector between the measured point and the second adjacent point;

[0102] The first collinear angle characteristic is determined based on the angle between the fifth and sixth vectors;

[0103] The second collinearity angle characteristic is determined based on the angle between the fifth and seventh vectors.

[0104] Optionally, in another possible implementation of the second aspect, whether the first collinear angle feature and the second collinear angle feature satisfy the collinearity condition is determined by the following method:

[0105] Determine whether the first collinear angle feature and the second collinear angle feature are less than the third angle threshold or greater than the fourth angle threshold;

[0106] If the first collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the first collinear angle feature satisfies the collinearity condition.

[0107] If the first collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the first collinear angle feature does not meet the collinearity condition.

[0108] If the second collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the second collinear angle feature satisfies the collinearity condition.

[0109] If the second collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the second collinear angle feature does not meet the collinearity condition.

[0110] Optionally, in another possible implementation of the second aspect, the reference window does not include adjacent points.

[0111] Thirdly, embodiments of this application provide a terminal device, including: 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 laser point cloud dragging point filtering method as described above.

[0112] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the laser point cloud dragging point filtering method as described above.

[0113] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the laser point cloud dragging and filtering method as described above.

[0114] The beneficial effects of this application embodiment compared with the prior art are as follows: by determining whether a point cloud point is a drag point based on the positional relationship between the point cloud point in the laser point cloud and its adjacent points in the horizontal direction and the reference window, and when a point cloud point is determined to be a drag point, the point cloud point is filtered out from the laser point cloud data, thereby effectively identifying and filtering drag points in the laser point cloud, thereby eliminating the impact of drag points on the recognition accuracy of the point cloud back-end algorithm, and improving the accuracy of back-end algorithms such as ranging and target recognition. Attached Figure Description

[0115] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0116] Figure 1This is a schematic flowchart of a laser point cloud dragging point filtering method provided in an embodiment of this application;

[0117] Figure 2 This is a schematic diagram of a portion of point cloud data in laser point cloud data provided in an embodiment of this application;

[0118] Figure 3 This is a schematic diagram of a reference window for determining the horizontal direction of a measured point according to an embodiment of this application;

[0119] Figure 4 This is a schematic diagram of another reference window for determining the measured point in the horizontal direction according to an embodiment of this application;

[0120] Figure 5 This is a schematic diagram of another reference window for determining the measured point in the horizontal direction according to an embodiment of this application;

[0121] Figure 6 This is a schematic diagram of another reference window for determining the measured point in the horizontal direction provided in an embodiment of this application;

[0122] Figure 7 This is a schematic diagram of another reference window for determining the measured point in the horizontal direction provided in an embodiment of this application;

[0123] Figure 8 This is a schematic diagram of another reference window for determining the measured point in the horizontal direction according to an embodiment of this application;

[0124] Figure 9 This is a schematic diagram of another reference window for determining the measured point in the horizontal direction according to an embodiment of this application;

[0125] Figure 10 This is a schematic diagram illustrating the positional relationship between the drag point and the front and rear reflective surfaces according to an embodiment of this application;

[0126] Figure 11 This is a schematic diagram illustrating the positional relationship between a drag point, adjacent points, and the origin, provided in an embodiment of this application.

[0127] Figure 12 This is a schematic diagram of adjacent points and a reference window in the horizontal direction of a measured point according to an embodiment of this application;

[0128] Figure 13 This is a flowchart illustrating a laser point cloud dragging point filtering method provided in another embodiment of this application;

[0129] Figure 14 This is a schematic diagram illustrating the positional relationship between a normal point and a reference window, provided in an embodiment of this application.

[0130] Figure 15 This is a schematic diagram illustrating the positional relationship between a drag point and a reference window, provided in an embodiment of this application.

[0131] Figure 16 This is a schematic diagram illustrating the positional relationship between a normal point, adjacent points, and the origin, provided in an embodiment of this application.

[0132] Figure 17 This is a schematic diagram illustrating the positional relationship between a drag point, adjacent points, and the origin, provided in an embodiment of this application.

[0133] Figure 18 This is a schematic diagram of the structure of the laser point cloud dragging point filtering device provided in the embodiments of this application;

[0134] Figure 19 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0135] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0136] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0137] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0138] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0139] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0140] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0141] The following description, with reference to the accompanying drawings, details the laser point cloud dragging and filtering method, apparatus, terminal equipment, storage medium, and computer program provided in this application.

[0142] Figure 1 The diagram shows a flowchart of a laser point cloud dragging point filtering method provided in an embodiment of this application.

[0143] like Figure 1 As shown, the method for filtering out drag points in the laser point cloud includes the following steps:

[0144] Step 101: Obtain the laser point cloud data to be processed collected by the lidar, wherein the laser point cloud data contains multiple point cloud points.

[0145] It should be noted that the laser point cloud dragging point filtering method of this application embodiment can be executed by the laser point cloud dragging point filtering device of this application embodiment. The laser point cloud dragging point filtering device of this application embodiment can be configured in any terminal device to execute the laser point cloud dragging point filtering method of this application embodiment. For example, the laser point cloud dragging point filtering device of this application embodiment can be configured in the on-board unit (OBU) of a vehicle equipped with a lidar to process the point cloud data collected by the lidar in the vehicle to remove dragging points in the laser point cloud, thereby improving the accuracy of subsequent ranging, target recognition and other algorithms, and improving vehicle driving safety.

[0146] The laser point cloud data to be processed can refer to the raw point cloud data collected by the lidar from the area being measured.

[0147] In this embodiment of the application, the lidar can collect laser point cloud data at a certain frequency. For example, if the lidar frequency is 100Hz, the lidar can collect 100 frames of laser point cloud data in 1 second. In this embodiment of the application, each frame of laser point cloud data collected by the lidar can be processed as laser point cloud data to be processed.

[0148] It should be noted that a lidar can emit one or more laser beams at a time and rotate with a certain horizontal angular resolution to generate a frame of laser point cloud data. For example, a lidar with a horizontal angular resolution of 0.1° and 128 vertical lines can generate a frame of laser point cloud data containing 360 / 0.1×128 = 460,800 point cloud points in each scan. Each point cloud point can be represented by a unique coordinate system, such as a Cartesian coordinate system, or by the lidar's horizontal angular resolution, the number of vertical lines, and the distance from the lidar.

[0149] As an example, the coordinates of point clouds in laser point cloud data can be represented by the horizontal angular resolution, the number of vertical lines, and the distance to the laser radar. Since each laser beam emitted by the laser radar generates a point cloud point during its rotation at horizontal angular resolution, and the distance between that point cloud point and the origin (the location of the laser radar) can be obtained, each point cloud point in the laser point cloud data can correspond to a number of vertical lines, a horizontal angle, and a distance to the origin. Therefore, these three parameters can be used as three-dimensional coordinates (number of lines, horizontal angle, distance) to represent the point cloud point. For example, for laser point cloud data collected by a laser radar with a horizontal angular resolution of 0.1° and 128 vertical lines, if the coordinates of a point cloud point are (64, 1°, 2m), it means that this point cloud point was collected by the 64th laser beam of the laser radar at a horizontal angle of 1° (during the 10th rotation), and the distance between this point cloud point and the origin is 2m.

[0150] It should be noted that in the following embodiments of this application, the coordinates of the point cloud points (number of lines, horizontal angle, distance) are used as an example for specific explanation.

[0151] Step 102: Determine at least one adjacent point of the measured point in the horizontal direction and at least one reference window.

[0152] The measured point can be any point cloud point in the laser point cloud data.

[0153] In this context, adjacent points in the horizontal direction of the measured point can refer to point cloud points with the same number of lines and adjacent horizontal angles as the measured point. For example, if the vertical line count of the lidar is 128 lines and the horizontal angular resolution is 0.1°, and the coordinates of the measured point are (60, 90°, x), then the adjacent points in the horizontal direction of the measured point can include point cloud points with 60 lines and a horizontal angle of 89.9°, as well as point cloud points with 60 lines and a horizontal angle of 90.1°. For edge points in the lidar point cloud data, there may only be one adjacent point in the horizontal direction. For example, assuming the coordinates of the measured point are (60, 0.1°, y), then the measured point has only one adjacent point with 60 lines and a horizontal angle of 0.2°; or assuming the coordinates of the measured point are (60, 360°, y), then the measured point has only one adjacent point with 60 lines and a horizontal angle of 359.9°.

[0154] In this embodiment of the application, each point in the laser point cloud data can be processed as a test point in sequence to determine whether each point is a drag point.

[0155] One possible approach is to determine all horizontally adjacent points of the measured point as its adjacent points. In other words, if the measured point is not an edge point (i.e., it has two horizontally adjacent points), then both of these adjacent points can be determined as its horizontal adjacent points and processed accordingly. If the measured point is an edge point (i.e., it has only one horizontal adjacent point), then that adjacent point can be determined as its horizontal adjacent point and processed accordingly.

[0156] For example, such as Figure 2 The diagram shows a partial view of point cloud data from a laser point cloud dataset. Assume the measured point is point cloud point 201 (this point cloud point is a dragged point), and it has two adjacent points in the horizontal direction: point cloud points 202 and 203. Therefore, point cloud points 202 and 203 can both be identified as adjacent points of the measured point in the horizontal direction. Assume the measured point is point cloud point 204 located at the edge of the point cloud data (this point cloud point is not a dragged point), and it has only one adjacent point in the horizontal direction: point cloud point 205. Therefore, point cloud point 205 can be identified as an adjacent point of the measured point in the horizontal direction.

[0157] As one possible implementation, any one of the horizontally adjacent points of the measured point can be selected for subsequent processing. That is, if the measured point is not an edge point, meaning it has two horizontally adjacent points, either of these two adjacent points can be randomly selected as the horizontally adjacent point of the measured point and processed accordingly. Alternatively, it can be pre-defined that adjacent points with a horizontal angle smaller than the horizontal angle of the measured point are selected as the horizontally adjacent points of the measured point and processed accordingly. Or, it can be pre-defined that adjacent points with a horizontal angle larger than the horizontal angle of the measured point are selected as the horizontally adjacent points of the measured point and processed accordingly.

[0158] For example, such as Figure 2 As shown, assuming the measured point is point cloud point 201, and it has two adjacent points in the horizontal direction: point cloud point 202 and point cloud point 203, then one of point cloud point 202 and point cloud point 203 can be selected as the adjacent point of the measured point in the horizontal direction. Assuming the measured point is point cloud point 204 located at the edge of the point cloud data, and it has only one adjacent point in the horizontal direction: point cloud point 205, then point cloud point 205 can be determined as the adjacent point of the measured point in the horizontal direction.

[0159] The reference window for the measured point in the horizontal direction can be composed of multiple point cloud points that are adjacent or close to the measured point in the horizontal direction. The reference window can contain multiple reference points; each reference point is a point cloud point in the laser point cloud data. For example, the number of reference points contained in the reference window may be 3, 5, etc., but this embodiment does not limit this.

[0160] As one possible implementation, the size of the reference window corresponding to the measured point can be fixed, meaning the number of reference points contained in the reference window can be fixed. Therefore, when the number of point cloud points on both sides of the measured point in the horizontal direction is greater than or equal to the size of the reference window, a reference window can be selected on both sides of the measured point, meaning there are two reference windows corresponding to the measured point; alternatively, a reference window can be selected on either side of the measured point. When the number of point cloud points on only one side of the measured point in the horizontal direction is greater than or equal to the size of the reference window, only one reference window can be selected on that side.

[0161] For example, such as Figure 3As shown in the figure, it is a schematic diagram of some point cloud points in the laser point cloud data. Assume that the size of the reference window is 5, and the measured point is point cloud point 301 (this point cloud point is a trailing point). It can be seen from the figure that the number of point cloud points on both sides of it (left and right in the figure) in the horizontal direction is greater than 5. Therefore, reference window 302 and reference window 303 can be determined as its two reference windows in the horizontal direction; alternatively, one of reference window 302 and reference window 303 can also be determined as its reference window in the horizontal direction.

[0162] For another example, as Figure 4 shown in the figure, assume that the measured point is point cloud point 401 at the edge of the point cloud data (this point cloud point is not a trailing point). It can be seen from the figure that there are no point cloud points on its left side, and the number of point cloud points on its right side is greater than 5. Therefore, reference window 402 can be determined as its reference window in the horizontal direction.

[0163] For another example, as Figure 5 shown in the figure, assume that the measured point is point cloud point 501. Since the number of point cloud points on its right side is 4, which is less than the size of the reference window, and the number of point cloud points on its left side is greater than 5, then reference window 502 can be determined as its reference window in the horizontal direction.

[0164] As a possible implementation, the size of the reference window corresponding to the measured point may not be fixed, that is, the number of reference points included in the reference window may not be fixed. For example, it can be set that the size of the reference window is greater than 1 and less than or equal to N, where N is a positive integer greater than 2. Therefore, when the number of point cloud points on both sides of the measured point in the horizontal direction is greater than or equal to N, a reference window with a size of N can be selected on both sides of the measured point, that is, the number of reference windows corresponding to the measured point is two and the sizes are both N; alternatively, a reference window with a size of N can also be selected on either side of the measured point.

[0165] Correspondingly, when only one side of the measured point in the horizontal direction has the number of point cloud points greater than or equal to N, and the number of point cloud points on the other side is M, where 1 < M < N, a reference window can be selected on both sides of the measured point, that is, the number of reference windows corresponding to the measured point is two, and one reference window has a size of N and the other reference window has a size of M; alternatively, a reference window can also be selected on either side of the measured point, such as selecting a reference window with a size of N or selecting a reference window with a size of M; or when it is set that one reference window can be selected, a reference window with a size of N can also be directly selected to improve the accuracy of trailing point recognition. "

[0166] Correspondingly, when only one side of the measured point in the horizontal direction has the number of point cloud points greater than or equal to N, only one reference window can be selected on that side.

[0167] For example, as Figure 3 As shown, assuming the size of the reference window is greater than 1 and less than or equal to 5, and the measured point is point cloud point 301 (this point cloud point is a drag point), it can be seen from the figure that the number of point cloud points on both sides (left and right sides in the figure) in the horizontal direction is greater than 5. Therefore, reference windows 302 and 303 can be determined as its two reference windows in the horizontal direction; or, one of reference windows 302 and 303 can be determined as its reference window in the horizontal direction.

[0168] For example, such as Figure 4 As shown in the figure, assuming the measured point is point cloud point 401 located at the edge of the point cloud data (this point cloud point is not a drag point), it can be seen from the figure that there are no point cloud points to its left and the number of point cloud points to its right is greater than 5. Therefore, reference window 402 can be determined as its reference window in the horizontal direction.

[0169] For example, such as Figure 6 As shown, assuming the measured point is point cloud point 601, since the number of point cloud points to its right is 4 and the number of point cloud points to its left is greater than 5, reference windows 602 and 603 can be determined as its reference windows in the horizontal direction, wherein the size of reference window 602 is 5 and the size of reference window 603 is 4; or, either reference window 602 or reference window 603 can be determined as its reference window in the horizontal direction; or, reference window 602 with a size of 5 can be directly set as its reference window in the horizontal direction.

[0170] It should be noted that, in Figures 3-6 In this context, each point cloud point within a reference window is a reference point corresponding to that reference window. Furthermore, when selecting adjacent points and reference windows, if the measured point has only one adjacent point in the horizontal direction and only one reference window can be selected, then one adjacent point and one reference window can be directly selected. For measured points with two adjacent points and two reference windows that can be selected, the user can choose to select one adjacent point and reference window or two adjacent points and reference windows according to actual needs and specific application scenarios; this application embodiment does not limit this choice.

[0171] As an example, for a test point that can select two adjacent points and a reference window, the two adjacent points and the reference window can be selected by default.

[0172] Furthermore, since the measured point is a drag point, its adjacent points or nearby point cloud points may also be drag points. Therefore, the presence of drag points in the reference window will affect the accuracy of drag point identification. Thus, a protection window can be set when selecting the reference window, and point cloud points outside the protection window can be used to construct the reference window to further improve the accuracy of drag point identification. That is, in one possible implementation of this application embodiment, the aforementioned adjacent points may not be included in the reference window.

[0173] As one possible implementation, when determining the reference window for the measured point in the horizontal direction, a protection window of size K can be set (i.e., the reference window contains K point cloud points, where K can be an integer greater than 1), and the protection window corresponding to the measured point can be centered on the measured point. Then, a reference window for the measured point in the horizontal direction can be determined outside the protection window, and the selection of one or two reference windows depends on the number of point cloud points outside the protection window.

[0174] For example, such as Figure 7 The diagram shows a partial view of point cloud data from a laser point cloud dataset. The size of the protection window is 5, and the size of the reference window is 5. Assuming the measured point is point cloud point 701, the position of its corresponding protection window is as follows: Figure 7 As shown, there are 5 point cloud points on the left side of the protection window and 5 point cloud points on the right side of the protection window. Therefore, a reference window can be determined on both the left and right sides of the protection window, namely reference window 1 and reference window 2. If it is preset that only one reference window is selected for each measured point, then either reference window 1 or reference window 2 can be determined as the reference window for that measured point in the horizontal direction.

[0175] For example, such as Figure 8 The diagram shows a partial view of point cloud data from a laser point cloud dataset. The size of the protection window is 5, and the size of the reference window is 5. Assuming the measured point is point cloud point 801, its corresponding protection window is as follows: Figure 8 As shown, since there are no point cloud points on the left side of the protection window and the number of point cloud points on the right side of the protection window is greater than 5, a reference window 1 can be determined on the right side of the protection window.

[0176] For example, such as Figure 9 The diagram shows a partial view of point cloud data from a laser point cloud dataset. The size of the protection window is 5, and the size of the reference window is 5. Assuming the measured point is point cloud point 901, its corresponding protection window is as follows: Figure 9 As shown, since the number of point cloud points on the right side of the protection window is less than 5 and the number of point cloud points on the left side of the protection window is greater than 5, a reference window 1 can be determined on the left side of the protection window.

[0177] It should be noted that when determining the reference window of the measured point in the horizontal direction, if a protection window is set, after excluding the point cloud points in the protection window, the reference window of the measured point in the horizontal direction can be determined in the same way as the determination of the reference window listed in the previous embodiment, which will not be repeated here.

[0178] As one possible implementation, since dragging points have a significant impact on near-range point cloud recognition, and the number of target points is smaller at long distances, no processing is required when the distance corresponding to a point cloud point is greater than a distance threshold; that is, dragging point recognition and filtering can be omitted for that point cloud point. In other words, the measured point in this embodiment can refer to a point cloud point whose corresponding distance is less than or equal to the distance threshold. It should be noted that the distance corresponding to a point cloud point can refer to the distance between the point cloud point and the origin, where the origin refers to the position of the LiDAR in the LiDAR point cloud data. In actual use, the specific value of the distance threshold can be determined according to actual needs and specific application scenarios; this embodiment does not limit this. For example, the distance threshold can be 50m.

[0179] Step 103: Determine the positional characteristics of the measured point relative to each adjacent point and each reference window.

[0180] It should be noted that since the dragged point is a point in the laser point cloud data located between two reflecting surfaces, its positional relationships, such as distances and angles, differ from those of the normal point cloud points on the two reflecting surfaces, including their proximity to adjacent points and their relationship with the normal point cloud points on the two reflecting surfaces. For example, the dragged point exhibits the following two characteristics in space:

[0181] 1. The drag point is usually located between the front and rear reflecting surfaces, and is not coplanar with either of the front or rear reflecting surfaces; for example... Figure 10 As shown, the angles (θ1 and θ2) between the drag point and the vectors corresponding to the two reflecting surfaces (reflecting surface 1 and reflecting surface 2) are usually not close to 0 or 180 degrees, and the distance corresponding to the drag point is usually between the two reflecting surfaces.

[0182] 2. According to the principle that light travels in straight lines, the line connecting the drag point to the edges of the two reflective surfaces usually passes through the origin (i.e., the location of the lidar). For example... Figure 11 As shown, the line connecting the drag point and the origin is usually collinear with the line connecting the left or right adjacent points.

[0183] Therefore, based on the above characteristics of drag points, the positional features between the measured point and each adjacent point and each reference window can be determined, and whether the measured point is a drag point can be determined based on whether the positional features between the measured point and each adjacent point and each reference window satisfy the characteristics corresponding to drag points.

[0184] As one possible implementation, the distance relationship between the measured point and adjacent points and the reflecting surface can be characterized by the distance between the measured point and adjacent points; the distance between the reflecting surface and the origin can be characterized by the distance between the point cloud points in the reference window and the origin; whether the measured point and adjacent points are collinear can be characterized by the angle between the measured point and adjacent points; and whether the measured point and the front and rear reflecting surfaces are coplanar can be characterized by the angle between the measured point and the reference window. Therefore, the above positional features can include distance features and angle features; correspondingly, step 103 can include:

[0185] Determine the distance characteristics between the measured point and each adjacent point, and the distance characteristics corresponding to each reference window;

[0186] Determine the angular characteristics between the measured point and each adjacent point and each reference window.

[0187] The distance feature between the measured point and its neighboring points may include the distance between the measured point and its neighboring point.

[0188] The distance features corresponding to the reference window may include at least one of the distance range and the distance mean of each reference point in the reference window.

[0189] Among them, the angular characteristics between the measured point and the adjacent point can refer to any collinear angular characteristics that can characterize whether the measured point and the adjacent point are collinear.

[0190] The angular feature between the measured point and the reference window can refer to any coplanar angular feature that can characterize whether the measured point and the reference point in the reference window are coplanar.

[0191] In this embodiment, as the foregoing analysis shows, since the lidar rotates with extremely low angular resolution during scanning, the distance between normally adjacent point cloud points in the horizontal direction should be very close. However, since the drag point is located between two reflective surfaces and is not on the same plane as its adjacent points, the distance between the drag point and its adjacent points should be relatively large. Therefore, when determining whether a measured point is a drag point, the distance between the drag point and its adjacent points can be used as a criterion for drag point identification, thereby determining the distance between the measured point and each adjacent point.

[0192] In addition, since the reference window contains multiple point cloud points, it can characterize the features of the reflecting surface to a certain extent. If the reflecting surface itself is uneven, the distance between normal adjacent point cloud points on the reflecting surface may also be large. Therefore, the range of distances between the reference points in the reference window can be used to determine whether the reflecting surface where the reference window is located is uneven. If the reflecting surface itself is uneven, even if the distance between the measured point and its adjacent points is large, it may be a normal point on the reflecting surface rather than a dragging point. Therefore, the range of distances between the reference points in the reference window can be used as a criterion for dragging point identification to prevent misjudgment of dragging points.

[0193] Furthermore, if two reference windows are identified in the horizontal direction for the measured point, and since the drag point is located between two reflective surfaces, if the measured point is a drag point, the point cloud points in these two reference windows are likely to be point cloud points on the two reflective surfaces, meaning the distance to the measured point falls between the average distances of these two reference windows. If the measured point is not a drag point, these two reference windows may both be point cloud points on a single reflective surface, and the distance to the measured point may be similar to the average distances of these two reference windows. Therefore, when identifying two reference windows in the horizontal direction for the measured point, the average distance of the reference points corresponding to each reference window can be used as a criterion for drag point identification to further improve the accuracy of drag point recognition.

[0194] In one possible implementation of this application embodiment, determining the distance features between the measured point and each adjacent point, and the distance features corresponding to each reference window, may include:

[0195] Determine the distance between the measured point and each adjacent point;

[0196] Determine the distance range corresponding to each reference point in each reference window.

[0197] As one possible implementation, for a neighboring point, the absolute value of the difference between the distance to the measured point and the distance to the neighboring point can be determined as the distance between the measured point and the neighboring point. Similarly, if multiple neighboring points are determined in step 102, the distance between the measured point and each neighboring point can be determined in the same way. For a reference window, the maximum and minimum distances corresponding to each reference point in the reference window can be determined, and the absolute value of the difference between the maximum and minimum distances corresponding to the reference window can be determined as the distance range corresponding to the reference window. Similarly, if multiple reference windows are determined in step 102, the distance range corresponding to each reference window can be determined in the same way.

[0198] As one possible implementation, if two reference windows in the horizontal direction of the measured point are determined in step 102, that is, the reference windows in the horizontal direction of the measured point include a first reference window of the measured point in the first horizontal direction and a second reference window of the measured point in the second horizontal direction, then the above determination of the distance features between the measured point and each adjacent point, and the distance features corresponding to each reference window, may further include:

[0199] Determine the mean first distance for each reference point in the first reference window;

[0200] Determine the mean of the second distance for each reference point in the second reference window.

[0201] In one possible implementation of this application embodiment, the average distance of each reference point in the first reference window can be determined as the average distance of the first reference window, and the average distance of each reference point in the second reference window can be determined as the average distance of the second reference window.

[0202] It should be noted that the first horizontal direction can refer to the direction in which the horizontal angle corresponding to the point cloud point gradually decreases, and the second horizontal direction can refer to the direction in which the horizontal angle corresponding to the point cloud point gradually increases; or, the first horizontal direction can refer to the direction in which the horizontal angle corresponding to the point cloud point gradually increases, and the second horizontal direction can refer to the direction in which the horizontal angle corresponding to the point cloud point gradually decreases. This application does not limit this specific direction in its embodiments. For example, as... Figure 12 As shown, assume the measured point is point cloud point 1201, and it has two adjacent points in the horizontal direction: point cloud point 1202 and point cloud point 1203. The first horizontal direction and the second horizontal direction are as follows: Figure 12 As shown, the first adjacent point can be point cloud point 1202, the second adjacent point can be 1203, the first reference window can be reference window 1, and the second reference window can be reference window 2.

[0203] In the embodiments of this application, based on the foregoing analysis and Figure 10 and Figure 11 It is known that the drag point is usually not coplanar with the front and rear reflecting surfaces, and the drag point is collinear with its adjacent points and the origin. In contrast, the normal point and its nearby point cloud points are usually coplanar, and the normal point is usually not collinear with its adjacent points and the origin. Therefore, whether the measured point is coplanar with each reference point in the reference window, and whether the measured point is collinear with its adjacent points and the origin, can be used as separate criteria for drag point identification. Thus, the coplanar angle characteristics between the measured point and each reference window, and the collinear angle characteristics between the measured point and each adjacent point, can be determined. That is, in one possible implementation of this application embodiment, the determination of the angle characteristics between the measured point and each adjacent point and each reference window can include:

[0204] Determine the coplanar angle characteristics between the measured point and each reference window. The coplanar angle characteristics can refer to the angle between the vector between the reference window and the measured point and the vector corresponding to the reference window.

[0205] Determine the collinear angle features between the measured point and each adjacent point. The collinear angle features can refer to the angle between the vector corresponding to the measured point and the origin and the vector corresponding to the measured point and the adjacent point. The origin refers to the position of the lidar in the lidar point cloud data.

[0206] As one possible implementation, for an adjacent point, the collinear angle characteristics between the measured point and the adjacent point can be determined based on the coordinates of the measured point, the coordinates of the adjacent point, and the coordinates of the origin. Similarly, if multiple adjacent points are determined in step 102, the collinear angle characteristics between the measured point and each adjacent point can be determined in the same way. For a reference window, the coplanar angle characteristics between the measured point and the reference window can be determined based on the coordinates of the measured point and the coordinates of each reference point within the reference window. Similarly, if multiple reference windows are determined in step 102, the coplanar angle characteristics between the measured point and each reference window can be determined in the same way.

[0207] As one possible implementation, if in step 102 two adjacent points and two reference windows in the horizontal direction of the measured point are determined, that is, the adjacent points in the horizontal direction of the measured point include the first adjacent point of the measured point in the first horizontal direction and the second adjacent point of the measured point in the second horizontal direction, and the reference windows in the horizontal direction of the measured point include the first reference window of the measured point in the first horizontal direction and the second reference window of the measured point in the second horizontal direction, wherein the first reference window contains multiple first reference points and the second reference window contains multiple second reference points, then the determination of the angular features between the measured point and each adjacent point and each reference window can include:

[0208] Determine the first coplanar angle characteristics between the measured point and the first reference window, and the second coplanar angle characteristics between the measured point and the second reference window;

[0209] Determine the first collinear angle characteristics between the measured point and the first adjacent point, and the second collinear angle characteristics between the measured point and the second adjacent point.

[0210] In one possible implementation of this application, the collinear angle characteristics between the measured point and the first adjacent point can be determined based on the coordinates of the measured point, the coordinates of the first adjacent point, and the coordinates of the origin; and the collinear angle characteristics between the measured point and the second adjacent point can be determined based on the coordinates of the measured point, the coordinates of the second adjacent point, and the coordinates of the origin. Furthermore, the coplanar angle characteristics between the measured point and the first reference window can be determined based on the coordinates of the measured point and the coordinates of each first reference point in the first reference window; and the coplanar angle characteristics between the measured point and the second reference window can be determined based on the coordinates of the measured point and the coordinates of each second reference point in the second reference window.

[0211] Step 104: Determine whether the measured point is a drag point based on the positional characteristics of the measured point with each adjacent point and each reference window.

[0212] As one possible implementation, the positional conditions corresponding to the drag point can be pre-configured. If the positional characteristics of the measured point with each adjacent point and each reference window satisfy the positional conditions corresponding to the drag point, the measured point can be determined as a drag point; if the positional characteristics of the measured point with each adjacent point and each reference window do not satisfy the positional conditions corresponding to the drag point, the measured point can be determined as a normal point.

[0213] As one possible implementation, a drag point recognition model can be pre-trained using a large amount of training data (a large number of point cloud point types (including dragged points and normal points) and their corresponding positional features). After determining the positional features of the measured point relative to each neighboring point and each reference window, these features can be input into the drag point recognition model. The model then identifies whether the measured point is a dragged point based on these positional features. If the output of the drag point recognition model indicates that the measured point is a dragged point, then it is confirmed as such; if the output indicates that the measured point is a normal point, then it is confirmed as such.

[0214] As one possible implementation, when the positional features between the measured point and its adjacent points and the reference window include distance features and angle features, it is also possible to determine whether each feature satisfies the feature corresponding to the drag point, so as to determine whether the measured point is a drag point.

[0215] As an example, the following method can be used to determine whether the measured point is a drag point:

[0216] If the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the distance range of each reference point in each reference window is less than the second distance threshold, and each coplanar angle feature does not meet the coplanar condition, then the measured point is determined to be a drag point. The coplanar condition is used to determine whether the measured point is coplanar with any reference window.

[0217] Otherwise, the measured point is determined to be a normal point;

[0218] or,

[0219] If the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the distance range of each reference point in each reference window is less than the second distance threshold, at least one coplanar angle feature satisfies the coplanar condition, and at least one collinear angle feature satisfies the collinear condition, then the measured point is determined to be a drag point. The collinear condition is used to determine whether the origin, the measured point and any adjacent point are collinear.

[0220] Otherwise, the measured point is determined to be a normal point.

[0221] As an example, if an adjacent point and a reference window are determined in step 102, the determination of whether the measured point is a drag point can be made by the following execution sequence:

[0222] Determine whether the distance between the measured point and its corresponding neighboring points is less than or equal to a first distance threshold.

[0223] If the distance between the measured point and its corresponding neighboring points is less than or equal to the first distance threshold, the measured point can be determined as a normal point.

[0224] If the distance between the measured point and its corresponding neighboring points is greater than the first distance threshold, it can be further determined whether the distance range of each reference point in the reference window is greater than or equal to the second distance threshold.

[0225] If the range of distances between each reference point in the reference window is greater than or equal to the second distance threshold, the measured point can be determined as a normal point.

[0226] If the range of distances between each reference point in the reference window is less than the second distance threshold, it is possible to continue to determine whether the coplanar angle features between the measured point and the reference window satisfy the coplanar condition.

[0227] If the coplanar angle features between the measured point and the reference window do not meet the coplanar condition, the measured point can be identified as a drag point.

[0228] If the coplanar angle features between the measured point and the reference window meet the coplanar condition, it is possible to further determine whether the collinear angle features between the measured point and adjacent points meet the collinear condition.

[0229] If the collinear angle characteristics between the measured point and its adjacent points satisfy the collinearity condition, the measured point can be identified as the drag point.

[0230] If the collinear angle characteristics between the measured point and its adjacent points do not meet the collinearity condition, the measured point can be determined as a normal point.

[0231] As an example, if two adjacent points and two reference windows are determined in step 102, the determination of whether the measured point is a drag point can be made by the following execution sequence:

[0232] Determine whether the distance between the measured point and each of its adjacent points is less than or equal to the first distance threshold;

[0233] If the distance between the measured point and its corresponding adjacent points is less than or equal to the first distance threshold, the measured point can be determined to be a normal point.

[0234] If the distance between the measured point and at least one adjacent point is greater than the first distance threshold, it can be further determined whether the distance range corresponding to each reference point in each reference window is greater than or equal to the second distance threshold.

[0235] If the distance range between each reference point in at least one reference window is greater than or equal to the second distance threshold, the measured point can be determined as a normal point.

[0236] If the range of distances between each reference point in each reference window is less than the second distance threshold, it is possible to further determine whether the coplanar angle features between the measured point and each reference window satisfy the coplanar condition.

[0237] If the coplanar angle features between the measured point and each reference window do not meet the coplanar condition, the measured point can be identified as a drag point.

[0238] If the coplanar angle features between the measured point and at least one reference window satisfy the coplanar condition, it is possible to further determine whether the collinear angle features between the measured point and each adjacent point satisfy the collinear condition.

[0239] If the collinear angular features between the measured point and at least one adjacent point satisfy the collinearity condition, the measured point can be identified as the drag point.

[0240] If the collinear angle characteristics between the measured point and each adjacent point do not meet the collinearity condition, the measured point can be determined as a normal point.

[0241] It should be noted that the number of adjacent points and reference windows, as well as the execution steps of each judgment condition, in the above examples are merely illustrative and should not be considered as limitations on this application. In actual use, the execution order of each judgment condition can be adjusted according to actual needs and specific application scenarios, and this application embodiment does not limit this.

[0242] Step 105: If the measured point is a drag point, remove the measured point from the laser point cloud data.

[0243] In this embodiment, if a measured point is determined to be a dragging point, it can be removed from the laser point cloud data; if a measured point is determined to be a normal point, it can be retained. After processing each point in the laser point cloud data, the processed laser point cloud data with dragging points removed can be obtained.

[0244] The drag point filtering method for laser point clouds provided in this application determines whether a point is a drag point based on the positional relationship between the point in the laser point cloud and its adjacent points in the horizontal direction and the reference window. When a point is determined to be a drag point, it is filtered out from the laser point cloud data. This effectively identifies and filters drag points in the laser point cloud, thereby eliminating the impact of drag points on the recognition accuracy of the point cloud backend algorithm and improving the accuracy of backend algorithms such as ranging and target recognition.

[0245] In one possible implementation of this application, dragging point identification can be performed on each point cloud in the laser point cloud data by using various discrimination conditions such as the distance relationship and positional relationship that the dragging point usually satisfies with adjacent points and the reflecting surface, so as to further improve the accuracy of dragging point identification and filtering.

[0246] The following is combined with Figure 13 The method for removing drag points from laser point clouds provided in the embodiments of this application will be further described.

[0247] Figure 13 This paper illustrates a flowchart of another laser point cloud dragging point filtering method provided in an embodiment of this application.

[0248] like Figure 13 As shown, the method for filtering out drag points in the laser point cloud includes the following steps:

[0249] Step 1301: Obtain the laser point cloud data to be processed collected by the lidar, wherein the laser point cloud data contains multiple point cloud points.

[0250] Step 1302: Determine the first adjacent point of the measured point in the first horizontal direction and the second adjacent point of the measured point in the second horizontal direction, as well as the first reference window of the measured point in the first horizontal direction and the second reference window of the measured point in the second horizontal direction.

[0251] The measured point can be any point cloud point in the laser point cloud data; the first reference window can contain multiple first reference points, and the second reference window can contain multiple second reference points.

[0252] Step 1303: Determine the first distance between the measured point and the first adjacent point, and the second distance between the measured point and the second adjacent point.

[0253] Step 1304: Determine the first distance range and the first distance mean for each first reference point in the first reference window.

[0254] Step 1305: Determine the second distance range and the second distance mean for each second reference point in the second reference window.

[0255] The specific implementation process and principle of steps 1301-1305 above can be found in the detailed description of the above embodiments, and will not be repeated here.

[0256] Step 1306: Determine the first coplanar angle feature between the measured point and the first reference window, and the second coplanar angle feature between the measured point and the second reference window.

[0257] In this embodiment, because the laser point cloud data is very dense, the distance between a point cloud point and some nearby point cloud points in actual space is very close. Therefore, since each reference point in the reference window is a point cloud point that is very close to the measured point in the horizontal direction, if the measured point is a normal point, then the measured point and each reference point should be roughly on a line. Therefore, according to... Figure 14 As shown, construct the reference window's vector α1 and the vector α2 between the reference window and the measured point. Then, the angle between vector α1 and vector α2 should be close to 0° or 180°. However, if the measured point is as shown... Figure 15 The drag point shown is not coplanar with the front and rear reflecting surfaces of reference window 1 and reference window 2, so if it is according to... Figure 15 In constructing vectors in this manner, angles θ1 and θ2 are typically not close to 0° or 180°. Therefore, as one possible implementation, step 1306 above could include:

[0258] Determine the first vector corresponding to the first reference window, and the second vector between the measured point and the first reference window;

[0259] The first coplanar angle feature is determined based on the angle between the first vector and the second vector;

[0260] Determine the third vector corresponding to the second reference window, and the fourth vector between the measured point and the second reference window;

[0261] The second coplanar angle feature is determined based on the angle between the third and fourth vectors.

[0262] One possible implementation is to first determine the first reference point with the largest distance from the measured point and the first reference point with the smallest distance from the measured point within the first reference window. Then, based on the coordinates of these two first reference points, determine a vector originating from the first reference point with the largest distance from the measured point and ending at the first reference point with the smallest distance from the measured point (or determine a vector originating from the first reference point with the smallest distance from the measured point and ending at the first reference point with the largest distance from the measured point), and define this vector as the first vector corresponding to the first reference window. Furthermore, based on the coordinates of the measured point and the coordinates of the first reference point with the largest distance from the measured point, determine a vector originating from the measured point and ending at the first reference point with the largest distance from the measured point (or determine a vector originating from the first reference point with the largest distance from the measured point and ending at the measured point), and define this vector as the second vector between the measured point and the first reference window. Finally, the angle between the first vector and the second vector is defined as the first coplanar angle feature between the measured point and the first reference window. Figure 15 For example, if the first reference window is reference window 1, then the first coplanar angle feature can be angle θ1.

[0263] Similarly, when determining the second coplanar angle feature, we can first determine the second reference point with the largest distance from the measured point and the second reference point with the smallest distance from the measured point within the second reference window; then, based on the coordinates of these two second reference points, we determine a vector that starts from the second reference point with the largest distance from the measured point and ends at the second reference point with the smallest distance from the measured point (or determine a vector that starts from the second reference point with the smallest distance from the measured point and ends at the second reference point with the largest distance from the measured point), and this vector is determined as the third vector corresponding to the second reference window; furthermore, based on the coordinates of the measured point and the coordinates of the second reference point with the largest distance from the measured point, we determine a vector that starts from the measured point and ends at the second reference point with the largest distance from the measured point (or determine a vector that starts from the second reference point with the largest distance from the measured point and ends at the measured point), and this vector is determined as the fourth vector between the measured point and the second reference window; finally, the angle between the third and fourth vectors is determined as the second coplanar angle feature between the measured point and the second reference window. Figure 15 For example, if the second reference window is reference window 1, then the second coplanar angle feature can be angle θ2.

[0264] Step 1307: Determine the first collinear angle feature between the measured point and the first adjacent point, and the second collinear angle feature between the measured point and the second adjacent point.

[0265] In the embodiments of this application, since the measured point, its horizontally adjacent points, and the origin are usually not collinear when the measured point is a normal point, there is usually a certain angle between the vector between the measured point and the origin and the vector between the measured point and its adjacent points; that is, the angle between these two is usually not close to 0° or 180°. Figure 16 As shown, the measured point is a normal point. Its vector to the origin is β1, its vector to adjacent point 1 is β2, and its vector to adjacent point 2 is β3. It is evident that the angles between β1 and β2, and between β1 and β3, are significantly not close to 0° and 180°, respectively. However, if the measured point is as follows... Figure 17 The vector between the measured point and the origin is β1, the vector between it and adjacent point 1 is β2, and the vector between it and adjacent point 2 is β3. Since the measured point is collinear with its horizontally adjacent points 1 and 2, the angles between β1 and β2 and between β1 and β3 are approximately 0° and 180°, respectively. Therefore, as a possible implementation, step 1307 above may include:

[0266] Determine the fifth vector between the measured point and the origin, where the origin refers to the position of the lidar in the lidar point cloud data;

[0267] Determine the sixth vector between the measured point and the first adjacent point, and the seventh vector between the measured point and the second adjacent point;

[0268] The first collinear angle characteristic is determined based on the angle between the fifth and sixth vectors;

[0269] The second collinearity angle characteristic is determined based on the angle between the fifth and seventh vectors.

[0270] As one possible implementation, a vector originating from the origin and ending at the measured point can be determined based on the coordinates of the measured point and the origin (or a vector originating from the measured point and ending at the origin), and this vector can be designated as the fifth vector between the measured point and the origin. Then, a vector originating from the measured point and ending at the first adjacent point can be determined based on the coordinates of the measured point and the first adjacent point (or a vector originating from the first adjacent point and ending at the measured point), and this vector can be designated as the sixth vector between the measured point and the first adjacent point. Vector; then, based on the coordinates of the measured point and the second adjacent point, a vector starting from the measured point and ending at the second adjacent point can be determined (or a vector starting from the second adjacent point and ending at the measured point can be determined), and this vector can be defined as the seventh vector between the measured point and the second adjacent point; finally, the angle between the fifth vector and the sixth vector can be defined as the first collinear angle feature between the measured point and the first adjacent point, and the angle between the fifth vector and the seventh vector can be defined as the second collinear angle feature between the measured point and the second adjacent point.

[0271] Step 1308: Determine whether the first distance and the second distance are both less than or equal to the first distance threshold. If yes, proceed to step 1313; otherwise, proceed to step 1309.

[0272] In this embodiment, since a drag point is usually located at a large distance from its horizontally adjacent points, when determining whether a measured point is a drag point, it can first be determined whether the distance between the measured point and its adjacent points satisfies the drag point characteristics. Therefore, it can first be determined whether the first distance between the measured point and the first adjacent point and the second distance between the measured point and the second adjacent point are both less than or equal to the first distance threshold. If both the first distance and the second distance are less than or equal to the first distance threshold, it can be determined that the distance between the measured point and its two adjacent points is small, thus determining that the measured point is a normal point. If the first distance is greater than the first distance threshold, or the second distance is greater than the first distance threshold, or both the first distance and the second distance are greater than the first distance threshold, it indicates that the distance between the measured point and at least one adjacent point is large, thus the measured point may be a drag point. In this case, the next step of judgment can be performed to further determine whether the measured point is a drag point through other judgment conditions, thereby improving the accuracy of drag point identification.

[0273] It should be noted that in actual use, the specific value of the first distance threshold can be determined according to actual needs and specific application scenarios, and this application embodiment does not limit this. For example, the first distance threshold can be 6cm.

[0274] Step 1309: Determine whether both the first distance range and the second distance range are less than the second distance threshold. If so, proceed to step 1310; otherwise, proceed to step 1313.

[0275] In this embodiment, since the distance between a normal point and its adjacent points may be large when the reflective surface is uneven, to prevent misjudgment of dragging points, when it is determined that the distance between the measured point and at least one adjacent point is large, it is further determined whether the reflective surface is uneven. Therefore, it can be determined whether the first distance range corresponding to the first reference window and the second distance range corresponding to the second reference window are both less than the second distance threshold. If both the first distance range and the second distance range are less than the second distance threshold, it can be said that the distances between each first reference point in the first reference window are relatively close, and the distances between each second reference point in the second reference window are also relatively close. Therefore, it can be determined that the reflective surfaces corresponding to the first and second reference windows are relatively flat, and the distances between adjacent point clouds should be relatively close. Therefore, in this case, since the distance between the measured point and its adjacent points is large, it can be determined that the measured point may be a dragging point. Therefore, the next step of judgment can be carried out to further determine whether the measured point is a dragging point through other judgment conditions, so as to improve the accuracy of dragging point identification. If at least one of the first distance range and the second distance range is greater than or equal to the second distance threshold, it indicates that the reflecting surface may be uneven, that is, the distance between the point cloud point on the reflecting surface and its adjacent points may be large. Therefore, although it is normal for the distance between the measured point and its adjacent points to be large, the measured point can be identified as a normal point.

[0276] It should be noted that in actual use, the specific value of the second distance threshold can be determined according to actual needs and specific application scenarios, and this application embodiment does not limit this. For example, the second distance threshold can be 10cm.

[0277] Step 1310: Determine whether the distance corresponding to the measured point is between the first average distance and the second average distance. If yes, proceed to step 1311; otherwise, proceed to step 1313.

[0278] In this embodiment, since the drag point is located between two reflective surfaces, if the distance corresponding to the measured point is between the average first distance of the first reference window and the average second distance of the second reference window, it can be determined that the measured point is between the two reflective surfaces. Therefore, it can be determined that the measured point may be a drag point, and further discrimination can be performed to further determine whether the measured point is a drag point through other discrimination conditions, thereby improving the accuracy of drag point identification. If the distance corresponding to the measured point is not between the average first distance and the average second distance, it can be determined that the measured point is on one of the reflective surfaces, thereby determining that the measured point is a normal point.

[0279] As one possible approach, if the distance corresponding to the measured point is greater than the smaller of the first and second average distances, and less than the larger of the first and second average distances, then the distance corresponding to the measured point can be determined to be between the first and second average distances. If the distance corresponding to the measured point is less than or equal to the smaller of the first and second average distances, or greater than or equal to the larger of the first and second average distances, then the distance corresponding to the measured point can be determined to be not between the first and second average distances.

[0280] As one possible implementation, since the distances between a normal point and its nearby points are close but not exactly the same, and the distance difference between the drag point and the normal points on the two reflective surfaces is usually large, if a measured point is located between the first and second average distance values ​​but is very close to either the first or second average distance value, it may also be a normal point. Therefore, to further improve the accuracy of drag point identification, the range of distances between the measured point and the front and rear reflective surfaces can be narrowed. That is, in one possible implementation of this application embodiment, step 1310 above may include:

[0281] The fourth distance threshold is determined based on the smaller of the first and second distance averages and the third distance threshold.

[0282] The fifth distance threshold is determined based on the larger of the first and second distance averages and the third distance threshold.

[0283] If the distance to the measured point is greater than the fourth distance threshold and less than the fifth distance threshold, it is determined that the distance to the measured point is between the first average distance and the second average distance.

[0284] If the distance to the measured point is less than or equal to the fourth distance threshold, or the distance to the measured point is greater than or equal to the fifth distance threshold, it is determined that the distance to the measured point is not between the first distance mean and the second distance mean.

[0285] As an example, the sum of the smaller of the first and second distance averages and a third distance threshold can be used to determine the fourth distance threshold; and the difference between the larger of the first and second distance averages and the third distance threshold can be used to determine the fifth distance threshold. If the distance corresponding to the measured point is greater than the fourth distance threshold and less than the fifth distance threshold, it can be determined that the distance corresponding to the measured point is between the first and second distance averages; if the distance corresponding to the measured point is less than or equal to the fourth distance threshold, or greater than or equal to the fifth distance threshold, it can be determined that the distance corresponding to the measured point is not between the first and second distance averages.

[0286] As an example, assume that the distance corresponding to the measured point is L, the first distance mean value corresponding to the first reference window is L1, the second distance mean value corresponding to the second reference window is L2, and the third distance threshold is δ. Assume that L1 < L2, then L1 + δ can be determined as the fourth distance threshold, and L2 - δ can be determined as the fifth distance threshold. If L1 + δ < L < L2 - δ, it can be determined that the distance corresponding to the measured point is between the first distance mean value and the second distance mean value; if L ≤ L1 + δ or L ≥ L2 - δ, it can be determined that the distance corresponding to the measured point is not between the first distance mean value and the second distance mean value.

[0287] It should be noted that in actual use, the specific value of the second distance threshold can be determined according to actual needs and specific application scenarios, and the embodiments of the present application do not limit this.

[0288] Step 1311: Determine whether at least one of the first coplanar angle feature and the second coplanar angle feature satisfies the coplanar condition. If so, execute Step 1312; otherwise, execute Step 1314.

[0289] In the embodiments of the present application, if the first coplanar angle feature between the measured point and the first reference window satisfies the coplanar condition, or the second coplanar angle feature between the measured point and the second reference window satisfies the coplanar condition, or both the first coplanar angle feature and the second coplanar angle feature satisfy the coplanar condition, it can be explained that the measured point is coplanar with the reflection surface where at least one reference window is located, and thus it may be a normal point. Therefore, the next step of discrimination can be continued to further determine whether the measured point is a drag point through other discrimination conditions to improve the accuracy of drag point recognition. If neither the first coplanar angle feature nor the second coplanar angle feature satisfies the coplanar condition, it can be explained that the measured point is not coplanar with the reflection surfaces where the first reference window and the second reference window are located, that is, the positional relationship between the measured point and the reference window satisfies Figure 15 the positional relationship shown, and thus it can be determined that the measured point is a drag point.

[0290] As a possible implementation manner, when the first coplanar angle feature is determined according to the angle between the aforementioned first vector and the second vector, and the second coplanar angle feature is determined according to the angle between the aforementioned third vector and the fourth vector, the following method can be used to determine whether the first coplanar angle feature and the second coplanar angle feature satisfy the coplanar condition:

[0291] Determine whether the first coplanar angle feature and the second coplanar angle feature are less than the first angle threshold or greater than the second angle threshold, where the second angle threshold is greater than the first angle threshold;

[0292] When the first coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, determine that the first coplanar angle feature satisfies the coplanar condition;

[0293] If the first coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the first coplanar angle feature does not meet the coplanar condition.

[0294] If the second coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the second coplanar angle feature satisfies the coplanar condition.

[0295] If the second coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the second coplanar angle feature does not satisfy the coplanar condition.

[0296] The first angle threshold is less than the second angle threshold. The first angle threshold is a smaller angle value close to 0°, and the second angle threshold is a larger angle value close to 180°.

[0297] As one possible implementation, if the first coplanar angle feature is the angle between the aforementioned first and second vectors, and the second coplanar angle feature is the angle between the aforementioned third and fourth vectors, that is... Figure 15 θ1 and θ2 are defined in the equation. Therefore, if the first coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it can be determined that the measured point and the first reference window are coplanar, thus confirming that the first coplanarity condition is met. If the first coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it can be determined that the measured point and the first reference window are not coplanar, thus confirming that the first coplanar angle feature does not meet the coplanarity condition. Similarly, if the second coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it can be determined that the measured point and the second reference window are coplanar, thus confirming that the second coplanarity condition is met. If the second coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it can be determined that the measured point and the second reference window are not coplanar, thus confirming that the second coplanar angle feature does not meet the coplanarity condition.

[0298] It should be noted that the first angle threshold is a small angle value close to 0°, and the second angle threshold is a large angle value close to 180°. In actual use, the specific values ​​of the first and second angle thresholds can be determined according to actual needs and specific application scenarios, and this application embodiment does not limit this. For example, the first angle threshold can be 20°, and the second angle threshold can be 160°.

[0299] Step 1312: Determine whether there is at least one collinearity condition among the first collinearity angle feature and the second collinearity angle feature. If yes, proceed to step 1314; otherwise, proceed to step 1313.

[0300] In this embodiment, if the first collinear angle feature between the measured point and the first adjacent point satisfies the collinearity condition, it can be said that the measured point, the first adjacent point, and the origin are collinear; if the second collinear angle feature between the measured point and the second adjacent point satisfies the collinearity condition, it can be said that the measured point, the second adjacent point, and the origin are collinear; if both the first and second collinear angle features satisfy the collinearity condition, it can be said that the measured point, the first adjacent point, and the origin are collinear, and the measured point, the second adjacent point, and the origin are also collinear. Figure 17 As shown; therefore, if at least one of the first collinear angle features and the second collinear angle feature satisfies the collinearity condition, the measured point can be determined to be a drag point. If neither the first collinear angle feature nor the second collinear angle feature satisfies the collinearity condition, it indicates that the measured point, the first adjacent point, and the origin are not collinear, and that the measured point, the second adjacent point, and the origin are not collinear, thus determining that the measured point is a normal point, such as... Figure 16 As shown.

[0301] As one possible implementation, when the first collinear angle feature is determined based on the angle between the aforementioned fifth and sixth vectors, and the second collinear angle feature is determined based on the angle between the aforementioned fifth and seventh vectors, it can be determined whether the first and second collinear angle features satisfy the collinearity condition in the following way:

[0302] Determine whether the first collinear angle feature and the second collinear angle feature are less than the third angle threshold or greater than the fourth angle threshold;

[0303] If the first collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the first collinear angle feature satisfies the collinearity condition.

[0304] If the first collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the first collinear angle feature does not meet the collinearity condition.

[0305] If the second collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the second collinear angle feature satisfies the collinearity condition.

[0306] If the second collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the second collinear angle feature does not meet the collinearity condition.

[0307] Among them, the third angle threshold is less than the fourth angle threshold. The third angle threshold is a smaller angle value close to 0°, and the fourth angle threshold is a larger angle value close to 180°.

[0308] As one possible implementation, when the first coplanar angle feature is the angle between the aforementioned fifth and sixth vectors, and the second coplanar angle feature is the angle between the aforementioned fifth and seventh vectors, if the first collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it can be determined that the measured point and the first adjacent point are collinear, thus confirming that the first collinearity condition is met. If the first collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it can be determined that the measured point and the first adjacent point are not collinear, thus confirming that the first collinear angle feature does not meet the collinearity condition. Similarly, if the second collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it can be determined that the measured point and the second adjacent point are collinear, thus confirming that the second collinearity condition is met. If the second collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it can be determined that the measured point and the second adjacent point are not collinear, thus confirming that the second collinear angle feature does not meet the collinearity condition.

[0309] It should be noted that the third angle threshold is a small angle value close to 0°, and the fourth angle threshold is a large angle value close to 180°. In actual use, the specific values ​​of the third and fourth angle thresholds can be determined according to actual needs and specific application scenarios, and this application embodiment does not limit this. For example, the first angle threshold can be 5°, and the second angle threshold can be 175°.

[0310] It should be noted that, in the embodiments of this application, when there are two adjacent points and two reference windows, if the first distance and / or the second distance is greater than the first distance threshold, the first distance range and the second distance range are both less than the second distance threshold, the distance corresponding to the measured point is between the first distance mean and the second distance mean, and the first coplanar angle feature and the second coplanar angle feature do not meet the coplanar condition, then the measured point can be determined as a drag point; otherwise, the measured point can be determined as a normal point. Alternatively, if the first distance and / or the second distance is greater than the first distance threshold, the first distance range and the second distance range are both less than the second distance threshold, the distance corresponding to the measured point is between the first distance mean and the second distance mean, the first coplanar angle feature and / or the second coplanar angle feature meet the coplanar condition, and the first collinear angle feature and / or the second collinear angle feature meet the collinear condition, then the measured point can be determined as a drag point; otherwise, the measured point can be determined as a normal point.

[0311] In other words, in this embodiment of the application, the execution steps of each judgment condition in steps 1308-1312 are not fixed. The execution order of steps 1308-1312 can be adjusted according to actual needs and specific application scenarios. This embodiment of the application does not limit this.

[0312] Step 1313: Determine that the measured point is a normal point.

[0313] Step 1314: Determine the measured point as the drag point.

[0314] Step 1315: If the measured point is a drag point, remove the measured point from the laser point cloud data.

[0315] The specific implementation process and principle of steps 1313-1315 above can be found in the detailed description of the foregoing embodiments, and will not be repeated here.

[0316] The drag point filtering method for laser point clouds provided in this application uses multiple discrimination conditions, such as whether the distance between the measured point and its adjacent points is large, whether the distance range of the measured point corresponding to the reference window in the horizontal direction is large, whether the measured point is between the average distances of the two reference windows, whether the measured point is coplanar with the reference window, and whether the measured point, its adjacent points, and the origin are collinear, to accurately identify whether the measured point is a drag point. This reduces the possibility of misjudging drag points and further improves the accuracy of drag point identification and filtering.

[0317] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0318] Corresponding to the laser point cloud dragging point filtering method described in the above embodiments, Figure 18 A schematic diagram of the structure of the laser point cloud dragging point filtering device provided in the embodiment of this application is shown. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0319] Reference Figure 18 The device 1800 includes:

[0320] The first acquisition module 1801 is used to acquire laser point cloud data to be processed collected by the lidar, wherein the laser point cloud data contains multiple point cloud points.

[0321] The first determining module 1802 is used to determine at least one adjacent point of the measured point in the horizontal direction and at least one reference window, wherein the measured point is any point cloud point in the laser point cloud data, and the reference window contains multiple reference points.

[0322] The second determining module 1803 is used to determine the positional characteristics of the measured point with each adjacent point and each reference window respectively;

[0323] The third determination module 1804 is used to determine whether the measured point is a drag point based on the positional characteristics of the measured point with each adjacent point and each reference window.

[0324] The first removal module 1805 is used to remove the measured point from the laser point cloud data when the measured point is a drag point.

[0325] The drag point filtering device for laser point clouds provided in this application determines whether a point cloud point is a drag point based on the positional relationship between the point cloud point and its adjacent points in the horizontal direction and the reference window. When a point cloud point is determined to be a drag point, it is filtered out from the laser point cloud data. This effectively identifies and filters drag points in the laser point cloud, thereby eliminating the impact of drag points on the recognition accuracy of the point cloud back-end algorithm and improving the accuracy of back-end algorithms such as ranging and target recognition.

[0326] In one possible implementation of this application, the aforementioned positional features include distance features and angle features; correspondingly, the aforementioned second determining module 1803 includes:

[0327] The first determining unit is used to determine the distance features between the measured point and each adjacent point, and the distance features corresponding to each reference window;

[0328] The second determining unit is used to determine the angular characteristics between the measured point and each adjacent point and each reference window.

[0329] Furthermore, in another possible implementation of this application, the aforementioned first determining unit is specifically used for:

[0330] Determine the distance between the measured point and each adjacent point;

[0331] Determine the distance range corresponding to each reference point in each reference window;

[0332] Accordingly, the second determining unit mentioned above is specifically used for:

[0333] Determine the coplanar angle characteristics between the measured point and each reference window, where the coplanar angle characteristics refer to the angle between the vector between the reference window and the measured point and the vector corresponding to the reference window;

[0334] Determine the collinear angle features between the measured point and each adjacent point. The collinear angle features refer to the angle between the vector corresponding to the measured point and the origin and the vector corresponding to the measured point and the adjacent point. The origin refers to the position of the lidar in the lidar point cloud data.

[0335] Furthermore, in another possible implementation of this application, the reference window includes a first reference window and a second reference window; correspondingly, the first determining unit is further configured to:

[0336] Determine the mean first distance for each reference point in the first reference window;

[0337] Determine the mean of the second distance for each reference point in the second reference window.

[0338] Furthermore, in another possible implementation of this application, the aforementioned third determining module 1804 includes:

[0339] The third determining unit is used to determine the measured point as a drag point when the distance between the measured point and at least one adjacent point is greater than the first distance threshold, the distance range corresponding to each reference point in each reference window is less than the second distance threshold, and each coplanar angle feature does not meet the coplanar condition. The coplanar condition is used to determine whether the measured point is coplanar with any reference window.

[0340] The fourth determining unit is used to determine the measured point as a normal point when any of the above conditions are not met;

[0341] or,

[0342] The fifth determining unit is used to determine the measured point as a drag point when the distance between the measured point and at least one adjacent point is greater than a first distance threshold, the distance range corresponding to each reference point in each reference window is less than a second distance threshold, at least one coplanar angle feature satisfies the coplanar condition, and at least one collinear angle feature satisfies the collinear condition. The collinear condition is used to determine whether the origin, the measured point and any adjacent point are collinear.

[0343] The sixth determining unit is used to determine the measured point as a normal point when none of the above conditions are met.

[0344] Furthermore, in another possible implementation of this application, the aforementioned third determining module 1804 includes:

[0345] The seventh determining unit is used to determine the measured point as a drag point if the distance between the measured point and at least one adjacent point is greater than a first distance threshold, the distance range corresponding to each reference point in each reference window is less than a second distance threshold, the distance corresponding to the measured point is between the first average distance and the second average distance, and each coplanar angle feature does not meet the coplanar condition. The coplanar condition is used to determine whether the measured point is coplanar with any reference window.

[0346] The eighth determining unit is used to determine the measured point as a normal point when any of the above conditions are not met;

[0347] or,

[0348] The ninth determining unit is used to determine the measured point as a drag point if the distance between the measured point and at least one adjacent point is greater than a first distance threshold, the distance range corresponding to each reference point in each reference window is less than a second distance threshold, the distance corresponding to the measured point is between the first distance mean and the second distance mean, at least one coplanar angle feature satisfies the coplanar condition, and at least one collinear angle feature satisfies the collinear condition. The collinear condition is used to determine whether the origin, the measured point and any adjacent point are collinear.

[0349] The tenth determining unit is used to determine the measured point as a normal point when none of the above conditions are met.

[0350] Furthermore, in another possible implementation of this application, the method for determining whether the distance corresponding to the measured point is between the first average distance and the second average distance is as follows:

[0351] The fourth distance threshold is determined based on the smaller of the first and second distance averages and the third distance threshold.

[0352] The fifth distance threshold is determined based on the larger of the first and second distance averages and the third distance threshold.

[0353] If the distance to the measured point is greater than the fourth distance threshold and less than the fifth distance threshold, it is determined that the distance to the measured point is between the first average distance and the second average distance.

[0354] If the distance to the measured point is less than or equal to the fourth distance threshold, or the distance to the measured point is greater than or equal to the fifth distance threshold, it is determined that the distance to the measured point is not between the first distance mean and the second distance mean.

[0355] Furthermore, in another possible implementation of this application, the aforementioned coplanar angle feature includes a first coplanar angle feature corresponding to the first reference window and a second coplanar angle feature corresponding to the second reference window; correspondingly, the aforementioned second determining unit is also used for:

[0356] Determine the first vector corresponding to the first reference window, and the second vector between the measured point and the first reference window;

[0357] The first coplanar angle feature is determined based on the angle between the first vector and the second vector;

[0358] Determine the third vector corresponding to the second reference window, and the fourth vector between the measured point and the second reference window;

[0359] The second coplanar angle feature is determined based on the angle between the third and fourth vectors.

[0360] Furthermore, in another possible implementation of this application, whether the first coplanar angle feature and the second coplanar angle feature satisfy the coplanar condition is determined by the following method:

[0361] Determine whether the first coplanar angle feature and the second coplanar angle feature are less than the first angle threshold or greater than the second angle threshold, wherein the second angle threshold is greater than the first angle threshold;

[0362] If the first coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the first coplanar angle feature satisfies the coplanar condition.

[0363] If the first coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the first coplanar angle feature does not meet the coplanar condition.

[0364] If the second coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the second coplanar angle feature satisfies the coplanar condition.

[0365] If the second coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the second coplanar angle feature does not satisfy the coplanar condition.

[0366] Furthermore, in another possible implementation of this application, the aforementioned collinear angle feature includes a first collinear angle feature corresponding to the first reference window and a second collinear angle feature corresponding to the second reference window; correspondingly, the aforementioned second determining unit is further used for:

[0367] Determine the fifth vector between the measured point and the origin;

[0368] Determine the sixth vector between the measured point and the first adjacent point, and the seventh vector between the measured point and the second adjacent point;

[0369] The first collinear angle characteristic is determined based on the angle between the fifth and sixth vectors;

[0370] The second collinearity angle characteristic is determined based on the angle between the fifth and seventh vectors.

[0371] Furthermore, in another possible implementation of this application, whether the first collinear angle feature and the second collinear angle feature satisfy the collinearity condition is determined by the following method:

[0372] Determine whether the first collinear angle feature and the second collinear angle feature are less than the third angle threshold or greater than the fourth angle threshold;

[0373] If the first collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the first collinear angle feature satisfies the collinearity condition.

[0374] If the first collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the first collinear angle feature does not meet the collinearity condition.

[0375] If the second collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the second collinear angle feature satisfies the collinearity condition.

[0376] If the second collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the second collinear angle feature does not meet the collinearity condition.

[0377] Furthermore, in another possible implementation of this application, the reference window does not include adjacent points.

[0378] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0379] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0380] To implement the above embodiments, this application also proposes a terminal device.

[0381] Figure 19 This is a schematic diagram of the structure of a terminal device according to an embodiment of this application.

[0382] like Figure 19 As shown, the terminal device 200 includes:

[0383] The system includes a memory 210 and at least one processor 220, and a bus 230 connecting different components (including the memory 210 and the processor 220). The memory 210 stores a computer program, which, when executed by the processor 220, implements the laser point cloud dragging filtering method described in this application embodiment.

[0384] Bus 230 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0385] Terminal device 200 typically includes various electronically readable media. These media can be any available media that can be accessed by terminal device 200, including volatile and non-volatile media, removable and non-removable media.

[0386] Memory 210 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 240 and / or cache memory 250. Terminal device 200 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 260 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 19 Not shown; usually referred to as a "hard drive"). Although Figure 19 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 230 via one or more data media interfaces. Memory 210 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0387] A program / utility 280 having a set (at least one) of program modules 270 may be stored in, for example, memory 210. Such program modules 270 include—but are not limited to—an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 270 typically perform the functions and / or methods described in the embodiments of this application.

[0388] Terminal device 200 can also communicate with one or more external devices 290 (e.g., keyboard, pointing device, display 291, etc.), and with one or more devices that enable a user to interact with terminal device 200, and / or with any device that enables terminal device 200 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 292. Furthermore, terminal device 200 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 293. As shown, network adapter 293 communicates with other modules of terminal device 200 via bus 230. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with terminal device 200, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0389] The processor 220 performs various functional applications and data processing by running programs stored in the memory 210.

[0390] It should be noted that the implementation process and technical principles of the terminal device in this embodiment are explained in the foregoing description of the laser point cloud dragging and filtering method in this application embodiment, and will not be repeated here.

[0391] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0392] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0393] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0394] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0395] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0396] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0397] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0398] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for filtering drag points in laser point clouds, characterized in that, include: Acquire laser point cloud data to be processed from lidar, wherein the laser point cloud data contains multiple point cloud points; Determine at least one adjacent point and at least one reference window in the horizontal direction of the measured point, wherein the measured point is any point cloud point in the laser point cloud data, and the reference window contains multiple reference points; Determine the positional features of the measured point relative to each of the adjacent points and each of the reference windows; Based on the positional characteristics of the measured point with each of the adjacent points and each of the reference windows, determine whether the measured point is a drag point; If the measured point is a drag point, the measured point is removed from the laser point cloud data.

2. The method as described in claim 1, characterized in that, The positional features include distance features and angle features. Determining the positional features between the measured point and each of the adjacent points and each of the reference windows includes: Determine the distance features between the measured point and each of the adjacent points, and the distance features corresponding to each of the reference windows; Determine the angular features between the measured point and each of the adjacent points and each of the reference windows.

3. The method as described in claim 2, characterized in that, Determining the distance features between the measured point and each of the adjacent points, and the distance features corresponding to each of the reference windows, includes: Determine the distance between the measured point and each of the adjacent points; Determine the distance range corresponding to each reference point in each reference window; Determining the angular features between the measured point and each of the adjacent points and each of the reference windows includes: Determine the coplanar angle feature between the measured point and each of the reference windows, wherein the coplanar angle feature refers to the angle between the vector between the reference window and the measured point and the vector corresponding to the reference window; Determine the collinear angle feature between the measured point and each of the adjacent points, wherein the collinear angle feature refers to the angle between the vector corresponding to the measured point and the origin and the vector corresponding to the measured point and the adjacent point, and the origin refers to the position of the lidar in the lidar point cloud data.

4. The method as described in claim 3, characterized in that, The reference window includes a first reference window and a second reference window. Determining the distance features between the measured point and each of the adjacent points, and the distance features corresponding to each reference window, further includes: Determine the average first distance corresponding to each of the reference points in the first reference window; Determine the average second distance for each reference point in the second reference window.

5. The method as described in claim 3 or 4, characterized in that, The step of determining whether a measured point is a drag point based on the positional characteristics of the measured point with each of the adjacent points and each of the reference windows includes: If the distance between the measured point and at least one of the adjacent points is greater than a first distance threshold, the distance range corresponding to each of the reference points in each of the reference windows is less than a second distance threshold, and each of the coplanar angle features does not satisfy the coplanar condition, then the measured point is determined to be a drag point, wherein the coplanar condition is used to determine whether the measured point is coplanar with any of the reference windows; Otherwise, the measured point is determined to be a normal point; or, If the distance between the measured point and at least one of the adjacent points is greater than the first distance threshold, the distance range corresponding to each of the reference points in each reference window is less than the second distance threshold, at least one of the coplanar angle features satisfies the coplanar condition, and at least one of the collinear angle features satisfies the collinear condition, then the measured point is determined to be a drag point. The collinear condition is used to determine whether the origin, the measured point and any of the adjacent points are collinear. Otherwise, the measured point is determined to be a normal point.

6. The method as described in claim 4, characterized in that, The step of determining whether a measured point is a drag point based on the positional characteristics of the measured point with each of the adjacent points and each of the reference windows includes: If the distance between the measured point and at least one of the adjacent points is greater than a first distance threshold, the distance range corresponding to each of the reference points in each reference window is less than a second distance threshold, the distance corresponding to the measured point is between the first average distance and the second average distance, and each of the coplanar angle features does not satisfy the coplanar condition, then the measured point is determined to be a drag point, wherein the coplanar condition is used to determine whether the measured point is coplanar with any of the reference windows; Otherwise, the measured point is determined to be a normal point; or, If the distance between the measured point and at least one of the adjacent points is greater than the first distance threshold, the distance range corresponding to each of the reference points in each reference window is less than the second distance threshold, the distance corresponding to the measured point is between the first average distance and the second average distance, at least one of the coplanar angle features satisfies the coplanar condition, and at least one of the collinear angle features satisfies the collinear condition, then the measured point is determined to be a drag point. The collinear condition is used to determine whether the origin, the measured point, and any of the adjacent points are collinear. Otherwise, the measured point is determined to be a normal point.

7. The method as described in claim 6, characterized in that, Whether the distance corresponding to the measured point is between the first average distance and the second average distance is determined by the following method: A fourth distance threshold is determined based on the smaller of the first average distance and the second average distance, and the third distance threshold. A fifth distance threshold is determined based on the larger of the first average distance and the second average distance, and the third distance threshold. If the distance corresponding to the measured point is greater than the fourth distance threshold and less than the fifth distance threshold, it is determined that the distance corresponding to the measured point is between the first average distance and the second average distance. If the distance corresponding to the measured point is less than or equal to the fourth distance threshold, or the distance corresponding to the measured point is greater than or equal to the fifth distance threshold, it is determined that the distance corresponding to the measured point is not between the first average distance and the second average distance.

8. The method as described in claim 6, characterized in that, The coplanar angle features include a first coplanar angle feature corresponding to the first reference window and a second coplanar angle feature corresponding to the second reference window. Determining the coplanar angle features between the measured point and each reference window includes: Determine the first vector corresponding to the first reference window, and the second vector between the measured point and the first reference window; The first coplanar angle feature is determined based on the angle between the first vector and the second vector; Determine the third vector corresponding to the second reference window, and the fourth vector between the measured point and the second reference window; The second coplanar angle feature is determined based on the angle between the third vector and the fourth vector.

9. The method as described in claim 8, characterized in that, The following method is used to determine whether the first coplanar angle feature and the second coplanar angle feature satisfy the coplanar condition: Determine whether the first coplanar angle feature and the second coplanar angle feature are less than a first angle threshold or greater than a second angle threshold, wherein the second angle threshold is greater than the first angle threshold; If the first coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the first coplanar angle feature satisfies the coplanar condition. If the first coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the first coplanar angle feature does not satisfy the coplanar condition. If the second coplanar angle feature is less than the first angle threshold or greater than the second angle threshold, it is determined that the second coplanar angle feature satisfies the coplanar condition; If the second coplanar angle feature is greater than or equal to the first angle threshold and less than or equal to the second angle threshold, it is determined that the second coplanar angle feature does not satisfy the coplanar condition.

10. The method as described in claim 6, characterized in that, The collinear angle feature includes a first collinear angle feature corresponding to the first reference window and a second collinear angle feature corresponding to the second reference window. Determining the collinear angle feature between the measured point and each of the adjacent points includes: Determine the fifth vector between the measured point and the origin; Determine the sixth vector between the measured point and the first adjacent point, and the seventh vector between the measured point and the second adjacent point; The first collinear angle feature is determined based on the angle between the fifth vector and the sixth vector; The second collinear angle feature is determined based on the angle between the fifth vector and the seventh vector.

11. The method as described in claim 10, characterized in that, The collinearity condition is determined by the following method: Determine whether the first collinear angle feature and the second collinear angle feature are less than the third angle threshold or greater than the fourth angle threshold; If the first collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the first collinear angle feature satisfies the collinearity condition. If the first collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the first collinear angle feature does not satisfy the collinearity condition. If the second collinear angle feature is less than the third angle threshold or greater than the fourth angle threshold, it is determined that the second collinear angle feature satisfies the collinearity condition. If the second collinear angle feature is greater than or equal to the third angle threshold and less than or equal to the fourth angle threshold, it is determined that the second collinear angle feature does not satisfy the collinearity condition.

12. The method according to any one of claims 1-11, characterized in that, The adjacent point is not included in the reference window.

13. A device for filtering out drag points in laser point clouds, characterized in that, include: The first acquisition module is used to acquire laser point cloud data to be processed collected by the lidar, wherein the laser point cloud data contains multiple point cloud points; The first determining module is used to determine at least one adjacent point of the measured point in the horizontal direction and at least one reference window, wherein the measured point is any point cloud point in the laser point cloud data, and the reference window contains multiple reference points; The second determining module is used to determine the positional features between the measured point and each of the adjacent points and each of the reference windows; The third determining module is used to determine whether the measured point is a drag point based on the positional characteristics between the measured point and each of the adjacent points and each of the reference windows. The first removal module is used to remove the measured point from the laser point cloud data when the measured point is a drag point.

14. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-12.

15. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-12.

Images