Tunnel noise filtering method and device based on vehicle-mounted radar, medium and radar system

Abstract

The invention provides a tunnel noisy point filtering method and device based on a vehicle-mounted radar, a medium and a radar system, relates to the technical field of point cloud processing, and aims to improve the quality of point cloud data in a tunnel, the method comprises the following steps: obtaining original point cloud data of a vehicle entering the tunnel, the original point cloud data comprising position information of a plurality of points; determining a tunnel contour point from the plurality of points based on the position information of the plurality of points; determining noisy points according to the position relationship between each point in the original point cloud data and the tunnel contour point; and screening out the noisy points from the original point cloud data to obtain processed point cloud data.

Description

Technical Field

[0001] The present application relates to the technical field of point cloud processing, and in particular to a tunnel noise filtering method, equipment, medium and radar system based on a vehicle-mounted radar. Background Art

[0002] Due to the multipath effect, the radar may generate false objects (noise points) with similar size and shape to some objects in the field of view outside the real position, especially objects with high reflectivity, such as traffic signs, cones, triangle signs, car license plates, taillights, etc. This phenomenon is particularly serious when the radar enters closed spaces such as tunnels. Since tunnel noise points may cause radar false alarms, affect the user's driving experience, and increase the user's potential driving risks, how to improve the quality of point cloud data in tunnels is a technical problem that needs to be solved urgently. Summary of the invention

[0003] In view of the above problems, the embodiments of the present application provide a tunnel noise filtering method, device, medium and radar system based on a vehicle-mounted radar, so as to overcome the above problems or at least partially solve the above problems.

[0004] In a first aspect of an embodiment of the present application, a method for filtering tunnel noise based on a vehicle-mounted radar is disclosed, the method comprising: Acquire original point cloud data of a vehicle entering a tunnel, wherein the original point cloud data includes position information of a plurality of points; Based on the position information of the plurality of points, determining a tunnel contour point from the plurality of points; Determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point; The noise points are filtered out from the original point cloud data to obtain processed point cloud data.

[0005] Optionally, determining the noise point according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: Determine points in the original point cloud data that are outside the tunnel contour points as noise points; The point outside the tunnel contour point is a point whose horizontal center axis distance of the vehicle is greater than the horizontal center axis distance of the vehicle to the tunnel contour point.

[0006] Optionally, determining the noise point according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: Determine at least one tunnel point located within the tunnel contour point in the original point cloud data as a noise point; The tunnel inner point is a point whose horizontal center axis distance of the vehicle is smaller than the horizontal center axis distance of the vehicle to the tunnel contour point.

[0007] Optionally, determining at least one tunnel point located within the tunnel contour point in the original point cloud data as a noise point comprises: The discreteness of each tunnel point located within the tunnel contour point in the original point cloud data is analyzed, and the tunnel points with a discreteness greater than a discreteness threshold are determined as noise points.

[0008] Optionally, analyzing the discreteness of each tunnel point located within the tunnel contour point in the original point cloud data, and determining the tunnel points with a discreteness greater than a discreteness threshold as noise points, comprises: Based on each tunnel point in the original point cloud data, traverse each adjacent point within a preset range centered on the tunnel point, and count a first total number of adjacent points within the preset range that satisfy a distance from the vertical center axis of the vehicle greater than or equal to a discrete point distance threshold; When the percentage of the first total number of points to the total number of points within the preset range is greater than or equal to the discrete point percentage threshold, the point centered at the preset range is determined as a noise point.

[0009] Optionally, the tunnel contour point includes a tunnel left edge point and a tunnel right edge point; and determining the tunnel contour point from the multiple points based on the position information of the multiple points includes: Traversing all points within the distance interval from the center point of the vehicle to the farthest detection point of the radar, and determining the point with the largest distance to the left of the horizontal center axis of the vehicle in the original point cloud data as the left edge point of the tunnel; And, the point with the largest distance to the right side of the horizontal center axis of the vehicle in the original point cloud data is determined as the right edge point of the tunnel.

[0010] Optionally, determining the noise point according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: Determine the points located outside the left edge point of the tunnel and the points located outside the right edge point of the tunnel in the original point cloud data as noise points; Wherein, the point outside the left edge point of the tunnel is a point whose left side distance from the horizontal center axis of the vehicle is greater than the distance from the horizontal center axis of the vehicle to the left edge point of the tunnel; The point outside the right edge point of the tunnel is a point whose right side distance from the horizontal center axis of the vehicle is greater than the distance from the horizontal center axis of the vehicle to the right edge point of the tunnel.

[0011] Optionally, the left edge point of the tunnel includes the left edge point of the tunnel in the first distance interval and the left edge point of the tunnel in the second distance interval; The right edge point of the tunnel includes the right edge point of the tunnel in the first distance interval and the right edge point of the tunnel in the second distance interval; The first distance interval is the distance from the center point of the vehicle to the first target detection point of the radar; The second distance interval is the distance from the first target detection point of the radar to the farthest detection point of the radar.

[0012] Optionally, the determining a tunnel contour point from the multiple points based on the position information of the multiple points includes: Traversing all points in the first distance interval, determining the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the first distance interval; And the point with the largest distance to the right side of the horizontal center axis of the vehicle is determined as the right edge point of the tunnel in the first distance interval.

[0013] Optionally, the determining a tunnel contour point from the multiple points based on the position information of the multiple points includes: Traversing all points in the second distance interval, and determining the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the second distance interval; And the point with the largest distance to the right side of the horizontal center axis of the vehicle is determined as the right edge point of the tunnel in the second distance interval.

[0014] Optionally, determining the noise point according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: For points whose distance from the vertical center axis of the vehicle is within the first distance interval, determining noise points according to the positional relationship between each point and the left edge point of the tunnel in the first distance interval or the right edge point of the tunnel in the first distance interval; For points whose distance from the vertical center axis of the vehicle is within the second distance interval, noise points are determined according to the positional relationship between each point and the left edge point of the tunnel in the second distance interval or the right edge point of the tunnel in the second distance interval.

[0015] Optionally, obtain the original point cloud data of the vehicle entering the tunnel, including: Traverse each point in the original point cloud data of the current frame, and count the second total number of points whose horizontal center axis distance of the vehicle is within the edge distance interval; When the percentage of the second total number of points in the total number of points of the original point cloud data of the current frame is greater than or equal to the tunnel percentage threshold, the original point cloud data of the current frame is determined as the original point cloud data of the vehicle entering the tunnel.

[0016] According to a second aspect of the embodiment of the present application, a tunnel noise filtering device based on a vehicle-mounted radar is disclosed, the device comprising: A first acquisition module is used to acquire original point cloud data of a vehicle entering a tunnel, wherein the original point cloud data includes position information of a plurality of points; A first determination module, configured to determine a tunnel contour point from the plurality of points based on position information of the plurality of points; A second determination module is used to determine the noise point according to the positional relationship between each point in the original point cloud data and the tunnel contour point; The first screening module is used to screen out the noise points from the original point cloud data to obtain processed point cloud data.

[0017] According to a third aspect of an embodiment of the present application, an electronic device is disclosed, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the tunnel noise filtering method based on the vehicle-mounted radar described in the first aspect of the embodiment of the present application are implemented.

[0018] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is disclosed, on which a computer program is stored. When the computer program is executed by a processor, the steps of the tunnel noise filtering method based on the vehicle-mounted radar described in the first aspect of the embodiment of the present application are implemented.

[0019] In a fifth aspect of the embodiments of the present application, a laser radar system is disclosed, comprising: the electronic device described in the third aspect of the embodiments of the present application, or the computer-readable storage medium described in the fourth aspect of the embodiments of the present application.

[0020] In the sixth aspect of the embodiments of the present application, a vehicle is disclosed, comprising: the electronic device described in the third aspect of the embodiments of the present application, or the computer-readable storage medium described in the fourth aspect of the embodiments of the present application, or the laser radar system described in the fifth aspect of the embodiments of the present application.

[0021] In a seventh aspect of an embodiment of the present application, a computer program product is disclosed, including a computer program, which, when executed by a processor, implements the steps of the tunnel noise filtering method based on a vehicle-mounted radar described in the first aspect of the embodiment of the present application.

[0022] The embodiments of the present application include the following advantages: In an embodiment of the present application, the original point cloud data of a vehicle entering a tunnel is obtained, and the original point cloud data includes the position information of multiple points; based on the position information of the multiple points, tunnel contour points are determined from the multiple points; noise points are determined according to the positional relationship between each point in the original point cloud data and the tunnel contour points; the noise points are filtered out from the original point cloud data to obtain processed point cloud data. Since the tunnel contour points are determined directly based on the positions of multiple points in the original point cloud data, there is no need to perform complex calculations. The tunnel contour points can be adaptively determined in a simple calculation method, and then the noise points in the original point cloud data can be quickly and effectively filtered out based on the tunnel contour points. In this way, a tunnel noise filtering method with a simple calculation method, fast and effective, is realized. This method can achieve high-accuracy and high-robustness tunnel noise filtering in a vehicle-mounted radar with limited hardware resources and high real-time requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0024] Figure 1 is a schematic diagram of a vehicle coordinate system provided in an embodiment of the present application; Figure 2 It is a flowchart of a method for filtering tunnel noise based on a vehicle-mounted radar provided in an embodiment of the present application; Figure 3 is a flowchart of another method for filtering tunnel noise based on a vehicle-mounted radar provided in an embodiment of the present application; Figure 4 is a flowchart of the steps of another method for filtering tunnel noise based on a vehicle-mounted radar provided in an embodiment of the present application; Figure 5 is a schematic diagram of a real tunnel scene provided by an embodiment of the present application; Figure 6 is a schematic diagram of a tunnel point cloud without noise filtering provided in an embodiment of the present application; Figure 7 This is a schematic diagram of a tunnel point cloud after noise filtering provided in an embodiment of the present application; Figure 8 It is a structural schematic diagram of a tunnel noise filtering device based on a vehicle-mounted radar provided in an embodiment of the present application; Fig. 9 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0026] In the related technology, the tunnel point cloud needs to be projected onto the elliptical tunnel cross section, and the background point cloud needs to be grid-projected and density-filtered. On this basis, a fast adaptive Euclidean clustering algorithm is used to filter the remaining scattered background noise points, and finally an adaptive elevation filtering threshold is used to delete invalid background point clouds such as the tunnel top. Therefore, the algorithm complexity of the existing noise filtering solutions is relatively high, and a relatively complex calculation method is required for noise filtering. There are problems such as high hardware resource requirements and poor real-time performance, which makes it difficult to apply to vehicle-mounted radars with limited hardware resources and high real-time requirements.

[0027] In order to overcome the limitations of related technologies, the embodiment of the present application provides a method for filtering tunnel noise based on a vehicle-mounted radar. The method directly determines the tunnel contour points based on the positions of multiple points in the original point cloud data, so there is no need to perform complex calculations. The tunnel contour points can be adaptively determined in a simple calculation method, and then the noise points in the original point cloud data can be quickly and effectively filtered out based on the tunnel contour points. A method for filtering tunnel noise with a simple calculation method and fast and effective is implemented. The method can achieve high-accuracy and high-robustness tunnel noise filtering in a vehicle-mounted radar with limited hardware resources and high real-time requirements.

[0028] The following is a detailed description of the tunnel noise filtering method based on vehicle-mounted radar provided in an embodiment of the present application in conjunction with the accompanying drawings.

[0029] In order to better understand the technical method implemented in the present application, the deployment of the vehicle-mounted radar and the vehicle coordinate system in the embodiment of the present application are first explained.

[0030] Reference Figure 1 As shown, Figure 1 It is a schematic diagram of a vehicle coordinate system provided in an embodiment of the present application. The vehicle-mounted radar is deployed on the top of the vehicle. The vehicle-mounted radar can be a laser radar or other radar. The embodiment of the present application does not limit the type of vehicle-mounted radar. The vehicle coordinate system is a three-dimensional rectangular coordinate system. The origin O of the vehicle coordinate system is located at the center point of the vehicle. The X-axis of the vehicle coordinate system coincides with the vertical center axis of the vehicle and points to the front of the vehicle; the Y-axis of the vehicle coordinate system coincides with the horizontal center axis of the vehicle and points to the side of the vehicle; the Z-axis of the vehicle coordinate system is perpendicular to the bottom surface and points to the top of the roof.

[0031] Reference Figure 2 As shown, Figure 2 1 is a flowchart of a method for filtering tunnel noise based on a vehicle-mounted radar provided in an embodiment of the present application. Figure 2 As shown, the tunnel noise filtering method based on the vehicle-mounted radar may include steps S210 to S240: Step S210: Acquire original point cloud data of a vehicle entering a tunnel, wherein the original point cloud data includes position information of a plurality of points.

[0032] Step S220: Based on the position information of the multiple points, determine the tunnel contour points from the multiple points.

[0033] Step S230: Determine noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point.

[0034] Step S240: filtering out the noise points from the original point cloud data to obtain processed point cloud data.

[0035] In the embodiment of the present application, the original point cloud data is the point cloud data collected by the vehicle-mounted radar for the environment where the current vehicle is located, and the original point cloud data contains multiple points (scanning points) at different positions. In the tunnel scene, all objects are inside the tunnel, and the points corresponding to the tunnel have specific position characteristics. Therefore, the tunnel contour points can be determined based on the position information of multiple points in the original point cloud data of the vehicle entering the tunnel. Among them, the position information of the point can reflect the spatial position of the point in the actual scene. In some embodiments, the position information includes the horizontal center axis distance of the vehicle (that is, the Y-axis distance value in the vehicle coordinate system) and the vertical center axis distance of the vehicle (that is, the X-axis distance value in the vehicle coordinate system).

[0036] Since all objects in the real tunnel scene are inside the tunnel, that is, all points in the original point cloud data should be located within the tunnel contour points, and there should be no points outside the tunnel contour points. Therefore, after determining the tunnel contour points, the noise points can be quickly determined based on the positional relationship between each point in the original point cloud data and the tunnel contour points, and the noise points can be filtered out from the original point cloud data to obtain the processed point cloud data.

[0037] In practical applications, the identifier of each point in the original point cloud data can be set to true (true) first, and after the noise point is determined through the above steps S210 to S230, the identifier of the noise point is set to false (false) in step S240, and finally the true (true) point of the identifier is output as the processed point cloud data.

[0038] In the above process, since the tunnel contour points are determined directly based on the positions of multiple points in the original point cloud data, there is no need for complex calculations. The tunnel contour points can be adaptively determined in a simple calculation method, and then the noise points in the original point cloud data can be quickly and effectively filtered out based on the tunnel contour points. In this way, a tunnel noise filtering method with a simple calculation method and fast and effective operation is realized. This method can achieve high accuracy and high robustness tunnel noise filtering in vehicle-mounted radars with limited hardware resources and high real-time requirements.

[0039] The following describes in detail the tunnel noise point filtering method based on vehicle-mounted radar in an embodiment of the present application through Section 1.1, which obtains the original point cloud data of the vehicle entering the tunnel, Section 1.2, which is a noise point determination method, and Section 1.3, which is a tunnel contour point determination method.

[0040] 1.1 Obtain the original point cloud data of the vehicle entering the tunnel: In combination with the above embodiments, in one implementation, the present application embodiment further provides a method for filtering tunnel noise based on a vehicle-mounted radar. In this method, the "obtaining original point cloud data of a vehicle entering a tunnel" in the above step 210 may specifically include sub-steps S210-1 to S210-2: Step S210 - 1 : traverse each point in the original point cloud data of the current frame, and count the second total number of points whose distance from the horizontal center axis of the vehicle is within the edge distance interval.

[0041] Step S210 - 2 : when the percentage of the second total number of points in the total number of points of the original point cloud data of the current frame is greater than or equal to the tunnel percentage threshold, the original point cloud data of the current frame is determined as the original point cloud data of the vehicle entering the tunnel.

[0042] In the embodiment of the present application, considering that the radar's field of view angle covers a larger area in the direction of the horizontal central axis (Y-axis) of the vehicle outside the tunnel, and the coverage area in the direction of the horizontal central axis of the vehicle decreases after entering the tunnel, it is possible to determine whether the current vehicle has entered the tunnel based on this feature, and then obtain the original point cloud data of the vehicle entering the tunnel.

[0043] Among them, the vehicle horizontal centerline distance refers to the Y-axis distance value in the vehicle coordinate system; the edge distance interval is a distance area set according to the tunnel cross-sectional size. The edge distance interval is an interval from the left edge threshold (y_left) to the right edge threshold (y_right) of the tunnel cross-section, that is, the edge distance interval is expressed as (y_left, y_right).

[0044] By traversing each point in the original point cloud data of the current frame, the total number of points (the second total number of points) of the vehicle's horizontal center axis distance within the edge distance interval is counted; for example, for the edge distance interval (y_left, y_right), the vehicle's horizontal center axis distance (y) in the original point cloud data of the current frame is counted within ( rght) total number of points.

[0045] Determine whether the percentage of the second total number of points in the total number of points in the original point cloud data of the current frame (i.e., the second total number of points divided by the total number of points in the original point cloud data of the current frame) is greater than or equal to the tunnel percentage threshold. If the percentage of the second total number of points in the total number of points in the original point cloud data of the current frame is greater than or equal to the tunnel percentage threshold, it means that the horizontal center axis distance of most points in the point cloud is in the edge distance interval, so it is determined that the vehicle has entered the tunnel scene. For example, if the tunnel percentage threshold is 80%, there are 100,000 points in the original point cloud data of the current frame, and if the horizontal center axis distance of 80,000 points is in the edge distance interval, then the percentage of the second total number of points in the total number of points in the point cloud is 80%, indicating that the vehicle has entered the tunnel.

[0046] In the above process, whether the vehicle enters the tunnel is determined based on the spatial distribution information of the original point cloud data, so as to obtain the original point cloud data of the vehicle entering the tunnel. There is no need to use complex algorithms to judge the scene in which the vehicle is located, which further reduces the computational complexity of the tunnel noise filtering method based on the vehicle-mounted radar, thereby realizing a simple, fast and effective tunnel noise filtering method.

[0047] 1.2 Noise determination method: According to the different locations of noise points, the noise points can be divided into noise points located inside the tunnel and noise points located outside the tunnel. The following describes the methods for determining the noise points located inside the tunnel and the noise points located outside the tunnel in the original point cloud data.

[0048] (1) Noise points outside the tunnel: In combination with the above embodiments, in one implementation, the present application embodiment further provides a method for filtering tunnel noise points based on a vehicle-mounted radar. In this method, the "determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point" in the above step S230 may specifically include: The points in the point cloud that are outside the tunnel contour point are determined as noise points; wherein the points that are outside the tunnel contour point are points whose vehicle horizontal center axis distance is greater than the vehicle horizontal center axis distance of the tunnel contour point.

[0049] In the embodiment of the present application, considering that all objects in the real tunnel scene are inside the tunnel, that is, all points in the original point cloud data should be located inside the tunnel contour points, and there should be no points outside the tunnel contour points, therefore, the points outside the tunnel contour points are determined as noise points.

[0050] Specifically, whether each point in the original point cloud data is outside the tunnel contour point is determined based on the relationship between the vehicle horizontal center axis distance of each point and the vehicle horizontal center axis distance of the tunnel contour point; wherein, the vehicle horizontal center axis distance refers to the Y-axis distance value in the vehicle coordinate system. Usually, the vehicle horizontal center axis distance of the tunnel contour point is the largest, so the point whose vehicle horizontal center axis distance is greater than the vehicle horizontal center axis distance of the tunnel contour point is determined as a noise point. For example, if the vehicle horizontal center axis distance (i.e., the Y-axis distance value) of the tunnel contour point is 5 meters, then the point whose Y-axis distance value in the original point cloud data is greater than 5 meters is determined as a noise point.

[0051] In the above process, the noise points outside the tunnel in the point cloud are simply and quickly determined based on the tunnel contour points, and these noise points can be filtered out without complicated calculations, thus realizing a simple, fast and effective method for filtering out noise points outside the tunnel.

[0052] (2) Noise points in the tunnel: In combination with the above embodiments, in one implementation, the present application embodiment further provides a method for filtering tunnel noise points based on a vehicle-mounted radar. In this method, the "determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point" in the above step S230 may specifically include: At least one in-tunnel point in the original point cloud data located within the tunnel contour point is determined as a noise point; wherein the in-tunnel point is a point whose vehicle horizontal center axis distance is smaller than the vehicle horizontal center axis distance of the tunnel contour point.

[0053] In the embodiment of the present application, the points whose horizontal center axis distance of the vehicle is less than the horizontal center axis distance of the vehicle to the tunnel contour point are regarded as points inside the tunnel. In addition to the points corresponding to real objects, there may also be noise points (i.e., noise points located inside the tunnel). For these noise points, a tunnel noise point filtering algorithm is executed to determine at least one tunnel point located inside the tunnel contour point in the original point cloud data as a noise point.

[0054] In a specific implementation, at least one in-tunnel point in the original point cloud data located within the tunnel contour point is determined as a noise point, including: analyzing the degree of discreteness of each in-tunnel point in the original point cloud data located within the tunnel contour point, and determining the in-tunnel point whose degree of discreteness is greater than a discreteness threshold as a noise point.

[0055] In the embodiment of the present application, considering that the points corresponding to real objects in the original point cloud data are relatively concentrated, the discreteness of each tunnel point located in the tunnel contour point in the original point cloud data is analyzed. The greater the discreteness, the more the point deviates from other tunnel points. The tunnel points with a discreteness greater than a discreteness threshold are then determined as noise points. The discreteness threshold is determined based on the actual application scenario and radar scanning accuracy.

[0056] Specifically, the discreteness of each tunnel point located within the tunnel contour point in the original point cloud data is analyzed, and the tunnel points with a discreteness greater than a discreteness threshold are determined as noise points, including steps A1 to A2: Step A1: Based on each tunnel point in the original point cloud data, traverse each adjacent point within a preset range centered on the tunnel point, and count the first total number of adjacent points within the preset range that meet the requirement that the distance from the vertical center axis of the vehicle is greater than or equal to the discrete point distance threshold.

[0057] Step A2: when the percentage of the first total number of points to the total number of points within the preset range is greater than or equal to the discrete point percentage threshold, the point centered at the preset range is determined as a noise point.

[0058] In the embodiment of the present application, the preset range refers to the area determined by a row multiplied by b columns with the point in the tunnel as the center, and the values ​​of a and b are flexibly set according to the actual situation. The discrete point distance threshold is determined based on the distance of the center point to the vertical center axis of the vehicle; for example, the distance of the center point to the vertical center axis of the vehicle is c, and the discrete point distance threshold (c+0.5) is determined based on the distance of the center point to the vertical center axis of the vehicle is c. If within the preset range, the distance of the vertical center axis of the vehicle of a certain adjacent point is (c+1), it means that the adjacent point deviates from the center point.

[0059] The total number of adjacent points (first total number of points) within the preset range that meet the requirement that the distance from the vertical center axis of the vehicle is greater than or equal to the discrete point distance threshold is counted. If the percentage of the first total number of points to the total number of points within the preset range (i.e., the first total number of points divided by the total number of points within the preset range) is greater than or equal to the discrete point percentage threshold, it indicates that the center point deviates from most points within the preset area, and thus the center point is determined to be a noise point. For example, if the discrete point percentage threshold is 60%, if the first total number of points accounts for 50% of the total number of points within the preset range, then the center point is not a noise point, and if the first total number of points accounts for 70% of the total number of points within the preset range, then the center point is a noise point.

[0060] Through the above process, the noise points in the tunnel and the points corresponding to the real objects can be effectively distinguished, so that the noise points located in the tunnel contour points are filtered out, thereby improving the accuracy of tunnel noise filtering.

[0061] 1.3 Tunnel contour point determination method: In combination with the above embodiments, in one implementation, the embodiment of the present application further provides a method for filtering tunnel noise points based on a vehicle-mounted radar. In this method, the tunnel contour points include tunnel left edge points and tunnel right edge points, and the "determining tunnel contour points from the multiple points based on the position information of the multiple points" in the above step S220 may specifically include: Traverse all points within the distance interval from the center point of the vehicle to the farthest detection point of the radar, and determine the point with the largest distance to the left of the horizontal central axis of the vehicle in the original point cloud data as the left edge point of the tunnel; and determine the point with the largest distance to the right of the horizontal central axis of the vehicle in the original point cloud data as the right edge point of the tunnel.

[0062] In the embodiment of the present application, in the tunnel scene, all objects are inside the tunnel. The tunnel contour points can be determined based on the distance of the horizontal center axis of the vehicle at each point in the original point cloud data. The point with the largest distance to the horizontal center axis of the vehicle is the tunnel contour point. The tunnel contour points include the left edge point of the tunnel and the right edge point of the tunnel. Therefore, the left edge of the tunnel is determined based on the point with the largest distance to the left of the horizontal center axis of the vehicle, and the right edge point of the tunnel is determined based on the point with the largest distance to the right of the horizontal center axis of the vehicle. Among them, the left edge point of the tunnel and the right edge point of the tunnel refer to the points detected by the radar, and are not the real tunnel contour points.

[0063] In the specific implementation, taking the center point of the vehicle as the starting point and the farthest detection point of the radar as the end point, all points in the distance interval from the center point of the vehicle to the farthest detection point of the radar are traversed along the vertical central axis (X-axis) of the vehicle to count the point with the largest distance to the left of the horizontal central axis of the vehicle (that is, the point with the largest distance value on the left Y-axis) and the point with the largest distance to the right of the horizontal central axis of the vehicle (that is, the point with the largest distance value on the right Y-axis), and then obtain the left edge point and the right edge point of the tunnel.

[0064] Furthermore, in the above step S230, “determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point” specifically includes: Determine the points located outside the left edge point of the tunnel and the points located outside the right edge point of the tunnel in the original point cloud data as noise points; The point outside the left edge point of the tunnel is a point whose left side distance from the horizontal center axis of the vehicle is greater than the distance from the horizontal center axis of the vehicle to the left edge point of the tunnel; the point outside the right edge point of the tunnel is a point whose right side distance from the horizontal center axis of the vehicle is greater than the distance from the horizontal center axis of the vehicle to the right edge point of the tunnel; In an embodiment of the present application, the original point cloud data is divided into points on the left and points on the right by the vertical center axis of the vehicle; for the points on the left, the noise points are determined based on the relationship between the left side distance of the vehicle's horizontal center axis and the distance between the vehicle's horizontal center axis and the left edge point of the tunnel; for the points on the right, the noise points are determined based on the relationship between the right side distance of the vehicle's horizontal center axis and the distance between the vehicle's horizontal center axis and the right edge point of the tunnel.

[0065] Through the above method, the left edge point of the tunnel and the left edge point of the tunnel are quickly determined based on the point with the largest distance to the left side of the horizontal central axis of the vehicle and the point with the largest distance to the right side of the horizontal central axis of the vehicle in the original point cloud data. Without the need for complex calculations, the tunnel contour points can be adaptively determined in a simple calculation method, thereby realizing fast and effective filtering of noise points in the point cloud based on the tunnel contour points.

[0066] In combination with the above embodiments, in one implementation, the embodiment of the present application also provides a method for filtering tunnel noise based on a vehicle-mounted radar. In this method, the left edge point of the tunnel includes the left edge point of the tunnel in the first distance interval and the left edge point of the tunnel in the second distance interval; the right edge point of the tunnel includes the right edge point of the tunnel in the first distance interval and the right edge point of the tunnel in the second distance interval; wherein, the first distance interval is the distance from the center point of the vehicle to the first target detection point of the radar; the second distance interval is the distance from the first target detection point of the radar to the farthest detection point of the radar.

[0067] Specifically, the “determining tunnel contour points from the multiple points based on the position information of the multiple points” in the above step S220 may specifically include: Step B1: Traverse all points within the first distance interval, and determine the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the first distance interval; and determine the point with the largest distance to the right of the horizontal center axis of the vehicle as the right edge point of the tunnel in the first distance interval.

[0068] Step B2: Traverse all points within the second distance interval, and determine the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the second distance interval; and determine the point with the largest distance to the right of the horizontal center axis of the vehicle as the right edge point of the tunnel in the second distance interval.

[0069] In the embodiment of the present application, considering that the tunnel may be curved, as the distance from the vertical center axis of the vehicle increases, the contour points on both sides of the tunnel may be different, that is, the tunnel contour points in the close distance interval close to the vehicle and the tunnel contour points in the long distance interval away from the vehicle may be different.

[0070] In order to obtain more accurate tunnel contour points, the distance interval from the center point of the vehicle to the farthest detection point of the radar is divided into two distance intervals, namely the first distance interval and the second distance interval, through the radar first target detection point; wherein, the radar first target detection point is determined according to the radar detection accuracy, and the radar farthest detection point refers to the farthest point that the radar can detect. For example, if the radar first target detection point is a point 20 meters away from the center point of the vehicle, and the radar farthest detection point is a point 60 meters away from the center point of the vehicle, then the first distance interval is a distance interval of 0 to 20 meters, and the second distance interval is a distance interval of 20 to 60 meters.

[0071] For different distance intervals, the left edge point and the right edge point of the tunnel corresponding to the distance interval are counted respectively. Specifically, for the first distance interval, all points in the distance interval from the center point of the vehicle to the first detection point of the radar (i.e., the points in the first distance interval where the vertical center axis of the vehicle is located) are traversed in the direction of the vertical center axis (X axis) of the vehicle to count the point with the largest distance to the left of the horizontal center axis of the vehicle (i.e., the point with the largest distance value on the left side of the Y axis) and the point with the largest distance to the right of the horizontal center axis of the vehicle (i.e., the point with the largest distance value on the right side of the Y axis), thereby obtaining the left edge point of the tunnel in the first distance interval and the right edge point of the tunnel in the first distance interval.

[0072] For the second distance interval, all points from the first target detection point of the radar to the farthest detection point of the radar (i.e., points within the second distance interval from the vertical center axis of the vehicle) are traversed in the direction of the vertical center axis (X-axis) of the vehicle to count the point with the largest distance to the left of the horizontal center axis of the vehicle (i.e., the point with the largest distance on the left Y-axis) and the point with the largest distance to the right of the horizontal center axis of the vehicle (i.e., the point with the largest distance on the right Y-axis), thereby obtaining the left edge point of the tunnel in the second distance interval and the right edge point of the tunnel in the second distance interval.

[0073] Through the above process, the distance interval from the center point of the vehicle to the farthest detection point of the radar is divided into different distance intervals for statistics, and the left edge point of the tunnel and the right edge point of the tunnel corresponding to each distance interval are obtained. Then, the noise point is determined based on the more accurate left edge point of the tunnel and the right edge point of the tunnel to achieve more accurate noise point filtering.

[0074] In one embodiment, in step B1, "traversing all points in the first distance interval, determining the point with the largest distance to the left of the horizontal central axis of the vehicle as the left edge point of the tunnel in the first distance interval; and determining the point with the largest distance to the right of the horizontal central axis of the vehicle as the right edge point of the tunnel in the first distance interval" specifically includes: Step B11: determining a first sub-distance interval from the first distance interval according to the first redundant distance value.

[0075] Step B12: Traverse all points in the first sub-distance interval, determine the point with the largest distance to the left side of the horizontal center axis of the vehicle as the left edge point of the tunnel in the first distance interval, and determine the point with the largest distance to the right side of the horizontal center axis of the vehicle as the right edge point of the tunnel in the first distance interval.

[0076] In an embodiment of the present application, in order to reduce the number of calculations and improve the real-time performance of the algorithm, a sub-distance interval (i.e., the first sub-distance interval) is determined within the first distance interval, so as to traverse based on the points within the sub-distance interval, and then quickly determine the left edge point of the tunnel and the right edge point of the tunnel in the first distance interval. In some embodiments, the first sub-distance interval is the distance from the center point of the vehicle to the midpoint of the first target detection point of the radar plus or minus the distance interval determined by the first redundant distance value. For example, if the first target detection point of the radar is distance, and the first redundant distance value is y1, then the first sub-distance interval is the distance interval from (distance / 2-y1) to (distance / 2+y1).

[0077] In this way, for the first distance interval from the center point of the vehicle to the first target detection point of the radar, it is only necessary to calculate the left edge point and the right edge point of the tunnel within a certain sub-distance interval, without calculating all points within the first distance interval, thereby effectively reducing the number of calculations and improving the real-time performance of the algorithm. This method can adaptively determine the size of the current tunnel cross-section and has good robustness. In addition, the first sub-distance interval is determined by the first redundant distance value, further improving the robustness of the algorithm and making the obtained tunnel contour points more accurate.

[0078] In one embodiment, in step B2, "traversing all points in the second distance interval, determining the point with the largest distance to the left of the horizontal central axis of the vehicle as the left edge point of the tunnel in the second distance interval; and determining the point with the largest distance to the right of the horizontal central axis of the vehicle as the right edge point of the tunnel in the second distance interval" specifically includes: Step B21: determining a second sub-distance interval from the second distance interval according to the second redundant distance value.

[0079] Step B22: Traverse all points in the second sub-distance interval, determine the point with the largest distance to the left side of the horizontal center axis of the vehicle as the left edge point of the tunnel in the second distance interval, and determine the point with the largest distance to the right side of the horizontal center axis of the vehicle as the right edge point of the tunnel in the second distance interval.

[0080] In an embodiment of the present application, in order to reduce the number of calculations and improve the real-time performance of the algorithm, a sub-distance interval (i.e., the second sub-distance interval) is determined within the second distance interval, so as to traverse based on the points within the sub-distance interval, and then quickly determine the left edge point of the tunnel and the right edge point of the tunnel in the second distance interval. In some embodiments, the second sub-distance interval is the distance between the midpoint of the radar first target detection point and the farthest detection point of the radar plus or minus the distance interval determined by the second redundant distance value. For example, the first target detection point of the radar is distance, the farthest detection point of the radar is detection_range, and the second redundant distance value is y2, then the second sub-distance interval is the distance interval from ((detection_range-distance) / 2-y2) to ((detection_range-distance) / 2+y2).

[0081] It can be understood that the first redundant distance value and the second redundant distance value may be the same or different.

[0082] In this way, for the second distance interval from the center point of the vehicle to the radar target detection point, it is only necessary to calculate the left edge point and the right edge point of the tunnel within a certain sub-distance interval, without calculating all points within the second distance interval, thereby effectively reducing the number of calculations and improving the real-time performance of the algorithm. This method can adaptively determine the size of the current tunnel cross-section and has good robustness. In addition, the second sub-distance interval is determined by the second redundant distance value, further improving the robustness of the algorithm and making the obtained tunnel contour points more accurate.

[0083] Furthermore, in the above step S230, “determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point” specifically includes: Step S230-1: For points within the first distance interval from the vertical center axis of the vehicle, noise points are determined according to the positional relationship between each point and the left edge point of the tunnel within the first distance interval or the right edge point of the tunnel within the first distance interval.

[0084] Step S230-2: For points whose distance from the vertical center axis of the vehicle is within the second distance interval, noise points are determined according to the positional relationship between each point and the left edge point of the tunnel in the second distance interval or the right edge point of the tunnel in the second distance interval.

[0085] In the embodiment of the present application, for different distance intervals, for the points whose distances to the vertical center axis of the vehicle are within the distance interval, noise points are determined based on the left edge points and right edge points of the tunnel corresponding to the distance interval. Specifically, the points whose distances to the vertical center axis of the vehicle are within the distance interval are divided into left points and right points by the vertical center axis of the vehicle; for the points on the left side, the noise points are determined based on the size relationship between the distance to the left side of the horizontal center axis of the vehicle and the distance to the left edge point of the tunnel by the horizontal center axis of the vehicle; for the points on the right side, the noise points are determined based on the size relationship between the distance to the right side of the horizontal center axis of the vehicle and the distance to the right edge point of the tunnel by the horizontal center axis of the vehicle.

[0086] Through the above process, the misjudgment of the midpoint of the point cloud due to the difference in tunnel contour points in different distance intervals is avoided, thereby improving the accuracy of tunnel noise point filtering.

[0087] The tunnel noise filtering method based on vehicle-mounted radar of the present application is described below in conjunction with two specific embodiments.

[0088] Embodiment 1: Reference Figure 3 As shown, Figure 3 : is a flowchart of another method for filtering tunnel noise based on a vehicle-mounted radar provided in an embodiment of the present application. The method for filtering tunnel noise based on a vehicle-mounted radar includes steps S310 to S360: Step S310: traverse each point in the original point cloud data of the current frame, and count the second total number of points whose distance from the horizontal center axis of the vehicle is within the edge distance interval.

[0089] Step S320: when the percentage of the second total number of points in the total number of points of the original point cloud data of the current frame is greater than or equal to the tunnel percentage threshold, the original point cloud data of the current frame is determined as the original point cloud data of the vehicle entering the tunnel.

[0090] Step S330: Based on the position information of the multiple points, determine the tunnel contour points from the multiple points.

[0091] Step S340: Determine the points in the original point cloud data that are outside the tunnel contour point as noise points; wherein the points that are outside the tunnel contour point are points whose vehicle horizontal center axis distance is greater than the vehicle horizontal center axis distance of the tunnel contour point.

[0092] Step S350: Determine at least one in-tunnel point in the original point cloud data that is within the tunnel contour point as a noise point; wherein the in-tunnel point is a point whose vehicle horizontal center axis distance is smaller than the vehicle horizontal center axis distance of the tunnel contour point.

[0093] Specifically, at least one in-tunnel point located within the tunnel contour point in the original point cloud data is determined as a noise point, including: analyzing the degree of discreteness of each in-tunnel point located within the tunnel contour point in the original point cloud data, and determining the in-tunnel point whose degree of discreteness is greater than a discreteness threshold as a noise point.

[0094] Step S360: filtering out the noise points from the original point cloud data to obtain processed point cloud data.

[0095] In the embodiment of the present application, the distribution information of the point cloud space in the original point cloud data is used to quickly judge the scene in which the vehicle is located in a simple calculation method, and the tunnel contour points are adaptively determined, so that the noise points are identified according to the tunnel contour points, thereby realizing a simple, fast and effective tunnel noise filtering method. This method can achieve high-accuracy and high-robustness tunnel noise filtering in a vehicle-mounted radar with limited hardware resources and high real-time requirements.

[0096] Embodiment 2: Reference Figure 4 As shown, Figure 4 4 is a flowchart of another method for filtering tunnel noise based on a vehicle-mounted radar provided in an embodiment of the present application. The method for filtering tunnel noise based on a vehicle-mounted radar includes steps S401 to S4013: Step S401: traverse each point in the original point cloud data of the current frame, and count the second total number of points whose distance from the horizontal center axis of the vehicle is within the edge distance interval.

[0097] Step S402: Determine whether the percentage of the second total number of points in the total number of points of the original point cloud data of the current frame is greater than or equal to a tunnel percentage threshold.

[0098] Step S403: when the percentage of the second total number of points in the total number of points of the original point cloud data of the current frame is greater than or equal to the tunnel percentage threshold, the original point cloud data of the current frame is determined as the original point cloud data of the vehicle entering the tunnel.

[0099] Step S404: traverse each point in the original point cloud data, and each point identifier is set to true by default.

[0100] Step S405: Traverse all points in the first distance interval, determine the point with the largest distance to the left of the horizontal central axis of the vehicle as the left edge point of the tunnel in the first distance interval; and determine the point with the largest distance to the right of the horizontal central axis of the vehicle as the right edge point of the tunnel in the first distance interval. Traverse all points in the second distance interval, determine the point with the largest distance to the left of the horizontal central axis of the vehicle as the left edge point of the tunnel in the second distance interval; and determine the point with the largest distance to the right of the horizontal central axis of the vehicle as the right edge point of the tunnel in the second distance interval.

[0101] Among them, the first distance interval is the distance from the center point of the vehicle to the first target detection point of the radar; the second distance interval is the distance from the first target detection point of the radar to the farthest detection point of the radar.

[0102] Step S406: traverse each point in the original point cloud data and execute a tunnel filtering algorithm.

[0103] Step S407: Execute the noise point filtering algorithm outside the tunnel: The left edge point of the tunnel located in the first distance interval and the right edge point of the tunnel located in the first distance interval are determined as noise points, and the identifier of the noise point is set to false.

[0104] The left edge point of the tunnel located in the second distance interval and the right edge point of the tunnel located in the second distance interval are determined as noise points, and the identifier of the noise point is set to false.

[0105] Step S408: Execute the noise point filtering algorithm in the tunnel: process the points in the cloud that are within the tunnel contour points, that is, process the tunnel points within the left edge points and the right edge points of the tunnel in the first distance interval, and the tunnel points within the left edge points and the right edge points of the tunnel in the second distance interval.

[0106] Step S409: Based on each tunnel point in the original point cloud data, traverse each adjacent point within a preset range centered on the tunnel point, and count the first total number of adjacent points within the preset range that meet the requirement that the distance from the vertical center axis of the vehicle is greater than or equal to the discrete point distance threshold; determine whether the percentage of the first total number of points in the total number of points within the preset range is greater than or equal to the discrete point percentage threshold.

[0107] Step S4010: when the percentage of the first total number of points to the total number of points in the preset range is greater than or equal to the discrete point percentage threshold, determining the point centered on the preset range as a noise point; and setting the identifier of the noise point to false.

[0108] Step S4011: determine whether the vehicle is in a normal outdoor scene (ie, not in a tunnel scene).

[0109] Step S4012: Execute normal outdoor point cloud processing algorithm.

[0110] Step S4013: Output the points whose identifiers are true to obtain processed point cloud data.

[0111] In the embodiment of the present application, since the tunnel contour points are determined directly based on the positions of multiple points in the original point cloud data, there is no need to perform complex calculations. The tunnel contour points can be adaptively determined in a simple calculation method, and then the noise points in the original point cloud data can be quickly and effectively filtered out based on the tunnel contour points. In this way, a tunnel noise point filtering method with a simple calculation method and fast and effective filtering is realized. This method can achieve high accuracy and high robustness tunnel noise point filtering in a vehicle-mounted radar with limited hardware resources and high real-time requirements.

[0112] Furthermore, in order to better illustrate the performance of the tunnel noise filtering method based on the vehicle-mounted radar in the embodiment of the present application, a test is performed in an actual environment. For example, Figure 5 is a schematic diagram of a real tunnel scene provided in an embodiment of the present application. Figure 6 This is a schematic diagram of a tunnel point cloud without noise filtering provided in an embodiment of the present application. It can be seen that there are a large number of noise points in the tunnel point cloud without noise filtering (for example, the points in the solid line box are noise points outside the tunnel). Figure 7 It is a schematic diagram of a tunnel point cloud after noise filtering provided in an embodiment of the present application. After noise filtering is performed based on the tunnel noise filtering method based on the vehicle-mounted radar provided in an embodiment of the present application, the noise in the point cloud can be removed.

[0113] It can be understood that this method can also be applied to noise filtering in other enclosed scenes. When applied to noise filtering in other enclosed scenes, the tunnel contour point mentioned in the embodiment of the present application refers to the edge point of the enclosed space. For example, the left edge point of the tunnel refers to the left edge point of the enclosed space, and the right edge point of the tunnel refers to the right edge point of the enclosed space. The present application also provides a tunnel noise filtering device based on a vehicle-mounted radar, referring to Figure 8 As shown, Figure 8 : is a schematic diagram of the structure of a tunnel noise filtering device based on a vehicle-mounted radar provided in an embodiment of the present application, the device comprising: A first acquisition module 810 is used to acquire original point cloud data of a vehicle entering a tunnel, wherein the original point cloud data includes position information of a plurality of points; A first determination module 820, configured to determine a tunnel contour point from the plurality of points based on the position information of the plurality of points; A second determination module 830, configured to determine noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point; The first screening module 840 is used to screen out the noise points from the original point cloud data to obtain processed point cloud data.

[0114] In an optional embodiment, the second determining module includes: The first determination submodule is used to determine the points in the original point cloud data that are located outside the tunnel contour point as noise points; wherein the points that are located outside the tunnel contour point are points whose horizontal center axis distance of the vehicle is greater than the horizontal center axis distance of the vehicle of the tunnel contour point.

[0115] In an optional embodiment, the second determining module includes: The second determination submodule is used to determine at least one tunnel point in the original point cloud data located within the tunnel contour point as a noise point; wherein the tunnel point is a point whose vehicle horizontal center axis distance is smaller than the vehicle horizontal center axis distance of the tunnel contour point.

[0116] In an optional embodiment, the second determination submodule is further used to analyze the degree of discreteness of each tunnel point located within the tunnel contour point in the original point cloud data, and determine the tunnel points with a degree of discreteness greater than a discreteness threshold as noise points.

[0117] In an optional embodiment, the second determining submodule includes: A first statistical module is used for traversing each adjacent point within a preset range centered on the tunnel point based on each tunnel point of the original point cloud data, and counting a first total number of adjacent points within the preset range that satisfy a distance from the vertical center axis of the vehicle greater than or equal to a discrete point distance threshold; The noise point determination submodule is used to determine the point centered on the preset range as a noise point when the percentage of the first total number of points to the total number of points in the preset range is greater than or equal to a discrete point percentage threshold.

[0118] In an optional embodiment, the tunnel contour points include a left edge point of the tunnel and a right edge point of the tunnel. The first determination module is further used to traverse all points within a distance interval from the center point of the vehicle to the farthest detection point of the radar, and determine the point with the largest distance to the left of the horizontal central axis of the vehicle in the original point cloud data as the left edge point of the tunnel; and determine the point with the largest distance to the right of the horizontal central axis of the vehicle in the original point cloud data as the right edge point of the tunnel.

[0119] In an optional embodiment, the second determining module includes: The third determination submodule is used to determine the points in the original point cloud data that are located outside the left edge point of the tunnel and the points that are located outside the right edge point of the tunnel as noise points; wherein the point located outside the left edge point of the tunnel is a point whose distance to the left side of the horizontal center axis of the vehicle is greater than the distance to the horizontal center axis of the vehicle from the left edge point of the tunnel, and the point located outside the right edge point of the tunnel is a point whose distance to the right side of the horizontal center axis of the vehicle is greater than the distance to the horizontal center axis of the vehicle from the right edge point of the tunnel.

[0120] In an optional embodiment, the left edge point of the tunnel includes the left edge point of the tunnel in the first distance interval and the left edge point of the tunnel in the second distance interval; the right edge point of the tunnel includes the right edge point of the tunnel in the first distance interval and the right edge point of the tunnel in the second distance interval; the first distance interval is the distance from the center point of the vehicle to the first target detection point of the radar; the second distance interval is the distance from the first target detection point of the radar to the farthest detection point of the radar.

[0121] In an optional embodiment, the first determining module includes: The fourth determination submodule is used to traverse all points within the first distance interval, and determine the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the first distance interval; and determine the point with the largest distance to the right of the horizontal center axis of the vehicle as the right edge point of the tunnel in the first distance interval.

[0122] In an optional embodiment, the first determining module includes: The fifth determination submodule is used to traverse all points within the second distance interval, and determine the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the second distance interval; and determine the point with the largest distance to the right of the horizontal center axis of the vehicle as the right edge point of the tunnel in the second distance interval.

[0123] In an optional embodiment, the second determining module includes: a sixth determination submodule, configured to determine, for points within the first distance interval from the vertical center axis of the vehicle, noise points according to a positional relationship between each point and a left edge point of the tunnel within the first distance interval or a right edge point of the tunnel within the first distance interval; The seventh determination submodule is used to determine the noise point for the points within the second distance interval from the vertical center axis of the vehicle according to the positional relationship between each point and the left edge point of the tunnel in the second distance interval or the right edge point of the tunnel in the second distance interval.

[0124] In an optional embodiment, the first acquisition module includes: The second statistical module is used to traverse each point in the original point cloud data of the current frame and count the second total number of points whose horizontal center axis distance of the vehicle is within the edge distance interval; The point cloud data determination module is used to determine the current frame original point cloud data as the original point cloud data of the vehicle entering the tunnel when the percentage of the second total number of points in the total number of points of the current frame original point cloud data is greater than or equal to the tunnel percentage threshold.

[0125] The present application also provides an electronic device, referring to Fig. 9 , Fig. 9 Schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Fig. 9 As shown, the electronic device 900 includes: a memory 910 and a processor 920. The memory 910 and the processor 920 are connected via a bus communication. A computer program is stored in the memory 910. The computer program can be run on the processor 920 to implement the steps of the tunnel noise filtering method based on the vehicle-mounted radar described in the embodiment of the present application.

[0126] The embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the tunnel noise filtering method based on the vehicle-mounted radar described in the embodiment of the present application are implemented.

[0127] An embodiment of the present application also provides a laser radar system, including: the electronic device described in the embodiment of the present application, or the computer-readable storage medium described in the embodiment of the present application.

[0128] An embodiment of the present application also provides a vehicle, comprising: the electronic device described in the embodiment of the present application, or the computer-readable storage medium described in the embodiment of the present application, or the laser radar system described in the embodiment of the present application.

[0129] The embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the tunnel noise filtering method based on a vehicle-mounted radar described in the embodiment of the present application.

[0130] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0131] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods and devices according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0132] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0134] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0135] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0136] The above is a detailed introduction to a tunnel noise filtering method, device, medium and radar system based on a vehicle-mounted radar provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for a general technician in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A tunnel noise filtering method based on vehicle-mounted radar, characterized in that: The method comprises: Acquire original point cloud data of a vehicle entering a tunnel, wherein the original point cloud data includes position information of a plurality of points; Based on the position information of the plurality of points, determining a tunnel contour point from the plurality of points; Determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point; The noise points are filtered out from the original point cloud data to obtain processed point cloud data.

2. The method according to claim 1, characterized in that Determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: Determine points in the original point cloud data that are outside the tunnel contour points as noise points; The point outside the tunnel contour point is a point whose horizontal center axis distance of the vehicle is greater than the horizontal center axis distance of the vehicle to the tunnel contour point.

3. The method according to claim 1, characterized in that: Determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: Determine at least one tunnel point located within the tunnel contour point in the original point cloud data as a noise point; The tunnel inner point is a point whose horizontal center axis distance of the vehicle is smaller than the horizontal center axis distance of the vehicle of the tunnel contour point.

4. The method according to claim 3, characterized in that Determining at least one tunnel point located within the tunnel contour point in the original point cloud data as a noise point comprises: The discreteness of each tunnel point located within the tunnel contour point in the original point cloud data is analyzed, and the tunnel points with a discreteness greater than a discreteness threshold are determined as noise points.

5. The method according to claim 4, characterized in that Analyzing the discreteness of each tunnel point located within the tunnel contour point in the original point cloud data, and determining the tunnel points with a discreteness greater than a discreteness threshold as noise points, including: Based on each tunnel point in the original point cloud data, traverse each adjacent point within a preset range centered on the tunnel point, and count a first total number of adjacent points within the preset range that satisfy a distance from the vertical center axis of the vehicle greater than or equal to a discrete point distance threshold; When the percentage of the first total number of points to the total number of points within the preset range is greater than or equal to the discrete point percentage threshold, the point centered at the preset range is determined as a noise point.

6. The method according to claim 1, characterized in that The tunnel contour points include a tunnel left edge point and a tunnel right edge point; and determining the tunnel contour points from the multiple points based on the position information of the multiple points includes: Traversing all points within the distance interval from the center point of the vehicle to the farthest detection point of the radar, and determining the point with the largest distance to the left of the horizontal center axis of the vehicle in the original point cloud data as the left edge point of the tunnel; And, the point with the largest distance to the right side of the horizontal center axis of the vehicle in the original point cloud data is determined as the right edge point of the tunnel.

7. The method according to claim 6, characterized in that Determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: Determine the points located outside the left edge point of the tunnel and the points located outside the right edge point of the tunnel in the original point cloud data as noise points; The point outside the left edge point of the tunnel is a point whose left side distance from the horizontal center axis of the vehicle is greater than the distance from the left edge point of the tunnel to the horizontal center axis of the vehicle; The point outside the right edge point of the tunnel is a point whose right side distance from the horizontal center axis of the vehicle is greater than the distance from the horizontal center axis of the vehicle to the right edge point of the tunnel.

8. The method according to claim 6, characterized in that The left edge point of the tunnel includes the left edge point of the tunnel in the first distance interval and the left edge point of the tunnel in the second distance interval; The right edge point of the tunnel includes the right edge point of the tunnel in the first distance interval and the right edge point of the tunnel in the second distance interval; The first distance interval is the distance from the center point of the vehicle to the first target detection point of the radar; The second distance interval is the distance from the first target detection point of the radar to the farthest detection point of the radar.

9. The method according to claim 8, characterized in that The step of determining tunnel contour points from the plurality of points based on the position information of the plurality of points comprises: Traversing all points in the first distance interval, determining the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the first distance interval; And the point with the largest distance to the right side of the horizontal center axis of the vehicle is determined as the right edge point of the tunnel in the first distance interval.

10. The method according to claim 8, characterized in that The step of determining tunnel contour points from the plurality of points based on the position information of the plurality of points comprises: Traversing all points in the second distance interval, and determining the point with the largest distance to the left of the horizontal center axis of the vehicle as the left edge point of the tunnel in the second distance interval; And the point with the largest distance to the right side of the horizontal center axis of the vehicle is determined as the right edge point of the tunnel in the second distance interval.

11. The method according to claim 8, characterized in that Determining noise points according to the positional relationship between each point in the original point cloud data and the tunnel contour point includes: For points whose distance from the vertical center axis of the vehicle is within the first distance interval, determining noise points according to the positional relationship between each point and the left edge point of the tunnel in the first distance interval or the right edge point of the tunnel in the first distance interval; For points whose distance from the vertical center axis of the vehicle is within the second distance interval, noise points are determined according to the positional relationship between each point and the left edge point of the tunnel in the second distance interval or the right edge point of the tunnel in the second distance interval.

12. The method according to any one of claims 1 to 11, characterized in that: Get the original point cloud data of the vehicle entering the tunnel, including: Traverse each point in the original point cloud data of the current frame, and count the second total number of points whose horizontal center axis distance of the vehicle is within the edge distance interval; When the percentage of the second total number of points in the total number of points of the original point cloud data of the current frame is greater than or equal to the tunnel percentage threshold, the original point cloud data of the current frame is determined as the original point cloud data of the vehicle entering the tunnel.

13. An electronic 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, the tunnel noise filtering method based on vehicle-mounted radar according to any one of claims 1 to 12 is implemented.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed on a computer, the computer is enabled to execute the tunnel noise filtering method based on a vehicle-mounted radar according to any one of claims 1 to 12.

15. A laser radar system, characterized in that: include: The electronic device of claim 13, or the computer-readable storage medium of claim 14.

16. A vehicle, characterized in that: include: The electronic device as claimed in claim 13, or the computer-readable storage medium as claimed in claim 14, or the laser radar system as claimed in claim 15.

17. A computer program product, characterized in that The computer program product includes computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the tunnel noise filtering method based on a vehicle-mounted radar according to any one of claims 1 to 12.

Images