CN-121995400-A - Point cloud processing method, electronic equipment, storage medium and vehicle

Abstract

The application discloses a point cloud processing method, electronic equipment, a storage medium and a vehicle, relates to the technical field of signal processing, and aims to solve the problem that false point cloud data exist in point cloud data of a laser radar in the related technology, so that the quality of the point cloud data is poor. The point cloud processing method comprises the steps of obtaining original point cloud data, determining object points and ground points from the points based on the information of the points, screening out information of invalid points from the original point cloud data to obtain processed point cloud data, wherein the invalid points are other points except the object points and the ground points.

Inventors

• ZHANG YALUN
• PAN LIHONG
• HUANG SHUO
• ZHAO WEIBING
• PENG LINA

Assignees

• 比亚迪股份有限公司

Dates

Publication Date

20260508

Application Date

20241101

Claims (19)

1. 1. A method of point cloud processing, the method comprising: acquiring original point cloud data, wherein the original point cloud data comprises information of a plurality of points; determining object points and ground points from the plurality of points based on the information of the plurality of points; And screening out information of invalid points from the original point cloud data to obtain processed point cloud data, wherein the invalid points are other points except the object point and the ground point in the plurality of points.
2. 2. The method of claim 1, wherein determining object points and ground points from the plurality of points based on the information of the plurality of points comprises: Determining the object point from the plurality of points based on the information of the plurality of points; and determining the ground point from non-object points based on the information of the points, wherein the non-object points are points except the object point.
3. 3. The method of claim 2, wherein the determining the object point from the plurality of points based on the information of the plurality of points comprises: And determining the target point meeting the local consistency condition as the object point based on the position information of the target point and the position information of the neighborhood point of the target point, wherein the target point is any one point of the plurality of points.
4. 4. The method of claim 3, wherein the local consistency condition comprises a distance between any one or more of the upper, lower, left, right, and neighbor points of the object point and the object point being less than or equal to a first preset distance threshold.
5. 5. The method of claim 3, wherein the local consistency condition comprises a number of valid points in a neighborhood of the preset size of the object point being greater than or equal to a preset number, wherein the valid points are neighborhood points in the neighborhood of the preset size of the object point where a distance between the valid points and the object point is less than a second preset distance threshold.
6. 6. The method of claim 2, wherein determining the ground point from the non-object points based on the information of the plurality of points comprises: And determining the non-object point meeting the ground coplanarity condition and the distance consistency condition as the ground point based on the position information of the non-object point and the position information of the neighborhood point of the non-object point.
7. 7. The method of claim 6, wherein the ground coplanar condition comprises at least one of a ground line distance condition, a tilt angle condition, and a grade angle condition.
8. 8. The method of claim 7, wherein the ground line distance condition includes a neighborhood point in a preset longitudinal neighborhood of the non-object point and a distance between the non-object point and a reference point, both being greater than a dead zone distance and less than an upper ground line limit distance, wherein the reference point is a reference point of position information of a plurality of points in the original point cloud data.
9. 9. The method of claim 7, wherein the tilt angle conditions include a first tilt angle and a second tilt angle that are both less than a first predetermined angle, and an absolute value of a difference between the first tilt angle and the second tilt angle is less than a second predetermined angle, wherein the first tilt angle is a tilt angle between the non-object point and a first neighboring point in a longitudinal direction, and the second tilt angle is a tilt angle between the non-object point and a second neighboring point in a longitudinal direction.
10. 10. The method of claim 7, wherein the grade angle condition includes a grade angle of the non-object point relative to a reference point being a reference point for position information of a plurality of points in the raw point cloud data being less than a third preset angle and an absolute value of a difference between the grade angle and the first inclination angle being less than a fourth preset angle.
11. 11. The method according to claim 10, wherein the method further comprises: Determining the ground height of the current frame based on the position information of the first line point in the original point cloud data of the current frame; A slope angle of the non-object point with respect to the reference point is determined based on the position information of the non-object point and the ground height of the current frame.
12. 12. The method of claim 11, wherein the location information comprises a height value, wherein determining the ground height of the current frame based on the location information of the top row point in the original point cloud data of the current frame comprises: and determining the average value of the height values of the first line points in the original point cloud data of the current frame as the ground height of the current frame under the condition that the average value of the height values of the first line points in the original point cloud data of the current frame is smaller than a preset height.
13. 13. The method of claim 11, wherein the location information comprises a height value, wherein determining the ground height of the current frame based on the location information of the top row point in the original point cloud data of the current frame comprises: and determining the ground height of the previous frame or the preset ground height as the ground height of the current frame under the condition that the average value of the height values of the first row points in the original point cloud data of the current frame is larger than or equal to the preset height.
14. 14. The method of claim 6, wherein the distance consistency condition includes that a neighborhood point in a preset lateral neighborhood of the non-object point and a distance between the non-object point and a reference point are both within a preset distance range, wherein the reference point is a reference point of position information of a plurality of points in the original point cloud data.
15. 15. An electronic device comprising a processor and a memory for storing instructions executable by the processor; Wherein the processor is configured to execute the instructions to implement the method of any one of claims 1 to 14.
16. 16. A computer readable storage medium having stored thereon computer instructions which, when run on a computer, cause the computer to perform the method of any of claims 1 to 14.
17. 17. A lidar system comprising the electronic device of claim 15, or the computer-readable storage medium of claim 16.
18. 18. A vehicle comprising the electronic device according to claim 15, or the computer-readable storage medium according to claim 16, or the lidar system according to claim 17.
19. 19. A computer program product comprising computer instructions which, when run on an electronic device, cause the electronic device to perform the method of any one of claims 1 to 14.

Description

Point cloud processing method, electronic equipment, storage medium and vehicle Technical Field The present application relates to the field of signal processing technologies, and in particular, to a point cloud processing method, an electronic device, a storage medium, and a vehicle. Background Along with the continuous development of technology, in the detection field, the laser radar has the advantages of high resolution, good concealment and strong anti-interference capability, and is widely applied to the fields of intelligent robots, unmanned vehicles and the like. However, in the process of detecting a real object by the laser radar, the laser radar transmits and receives signals under the influence of environmental light, space scattered matters and the like. False point cloud data, namely noise points, appear in the point cloud data output by the laser radar, so that the quality of the point cloud data is poor, and the processing result is inaccurate. Therefore, how to improve the quality of the point cloud data is a technical problem to be solved at present. Disclosure of Invention The embodiment of the application aims to provide a point cloud processing method, electronic equipment, a storage medium and a vehicle, and aims to solve the problem that the quality of point cloud data is poor due to false point cloud data in the point cloud data of a laser radar in the related technology. In order to achieve the above purpose, the embodiment of the application adopts the following technical scheme: in a first aspect, an embodiment of the present application provides a point cloud processing method, where the method includes: acquiring original point cloud data, wherein the original point cloud data comprises information of a plurality of points; determining object points and ground points from the plurality of points based on the information of the plurality of points; and screening out information of invalid points from the original point cloud data to obtain processed point cloud data, wherein the invalid points are points except object points and ground points in a plurality of points. According to the point cloud processing method provided by the embodiment of the application, based on the information of a plurality of points in the original point cloud data, the object points and the ground points of the real object are determined from the original point cloud data, and invalid points except the object points and the ground points in the original point cloud data are screened out, so that false point cloud data in the processed point cloud data are avoided, and the quality of the point cloud data is improved. In some embodiments, determining the object point and the ground point from the plurality of points based on the information of the plurality of points includes determining the object point from the plurality of points based on the information of the plurality of points and determining the ground point from the non-object point based on the information of the plurality of points, the non-object point being other points of the plurality of points than the object point. In some embodiments, determining the object point from the plurality of points based on the information of the plurality of points includes determining the target point satisfying the local consistency condition as the object point based on the position information of the target point and the position information of a neighborhood point of the target point, the target point being any one of the plurality of points. In some embodiments, the local consistency condition includes that a distance between any one or more of the upper, lower, left and right neighborhood points of the object point and the object point is less than or equal to a first preset distance threshold. In some embodiments, the local consistency condition includes that the number of effective points in a neighborhood of the preset size of the object point is greater than or equal to the preset number, wherein the effective points are neighborhood points of which the distance between the neighborhood of the preset size of the object point and the object point is smaller than a second preset distance threshold. In some embodiments, determining the ground point from the non-object points based on the information of the plurality of points includes determining the non-object points satisfying the ground coplanarity condition and the distance consistency condition as the ground points based on the position information of the non-object points and the position information of the neighbor points of the non-object points. In some embodiments, the ground coplanar condition includes at least one of a ground line distance condition, a tilt angle condition, and a grade angle condition. In some embodiments, the ground line distance condition comprises that the distances between the neighborhood point in the preset longitudinal neighborhood of the non-object point and the re