LiDAR-based object detection method and apparatus
View Patent ↗A LiDAR-based object detection method includes clustering a point cloud acquired from LiDAR, selecting a to-be-divided cluster among clusters generated in the clustering, and selecting division points according to a geometrical feature formed with adjacent points from among points belonging to the to-be-divided cluster, and dividing the to-be-divided cluster based on a representative point determined by at least some of the division points.
1. A LiDAR-based object detection method comprising:
obtaining a point cloud for surrounding environment by use of LiDAR;
clustering, by a microprocessor, points of the point cloud acquired from the LiDAR;
selecting, by the microprocessor, a to-be-divided cluster among clusters generated in the clustering; and
selecting, by the microprocessor, division points according to a geometrical feature formed with adjacent points from among points belonging to the to-be-divided cluster, and dividing the to-be-divided cluster based on a representative point determined by at least some of the division points.
2. The LiDAR-based object detection method according to claim 1 , wherein the dividing includes:
selecting division points according to the geometrical feature among points belonging to the to-be-divided cluster;
clustering the division points; and
selecting a final division point cluster among clusters of the division points, and dividing the to-be-divided cluster using a representative point of the final division point cluster.
3. The LiDAR-based object detection method according to claim 2 , wherein the dividing is performed based on a straight line passing through the representative point.
4. The LiDAR-based object detection method according to claim 3 , wherein the straight line is obtained by connecting the representative point and an origin.
5. The LiDAR-based object detection method according to claim 2 , wherein the dividing is performed when a ratio of a number of points in at least one cluster divided from the to-be-divided cluster to a total number of points in the to-be-divided cluster is less than or equal to a reference value.
6. The LiDAR-based object detection method according to claim 2 , wherein one of two division point clusters is selected as the final division point cluster, the two division point clusters having representative points forming maximum and minimum angles, respectively, with a reference line passing through an origin.
7. The LiDAR-based object detection method according to claim 6 , wherein the final division point cluster is selected based on a difference in a number of points between two divided clusters when the to-be-divided cluster divided by a feature angle thereof.
8. The LiDAR-based object detection method according to claim 2 , wherein, after mapping the division points to a 2D grid map, the clustering of the division points is performed based on a degree of proximity of grids.
9. The LiDAR-based object detection method according to claim 8 , wherein grids located as directly connected to each other among the grids including the division points are determined as the same division point cluster.
10. The LiDAR-based object detection method according to claim 1 , wherein the representative point is determined by average coordinate values of at least some of the division points.
11. The LiDAR-based object detection method according to claim 1 , wherein the to-be-divided cluster is selected according to a number of division points therein.
12. The LiDAR-based object detection method according to claim 1 , wherein the geometrical feature includes an angle θ formed with the adjacent points.
13. The LiDAR-based object detection method according to claim 12 , wherein the division points are selected according to whether the angle θ is an acute angle.
14. The LiDAR-based object detection method according to claim 13 , wherein whether the angle θ is an acute angle is determined using “1−cos θ”.
15. The LiDAR-based object detection method according to claim 12 , wherein the adjacent points are points on the same layer as the one of the division points.
16. A LiDAR-based object detection apparatus comprising:
a microprocessor;
a memory; and
an input/output device,
wherein the microprocessor executes the method according to claim 1 .
17. A vehicle comprising the LiDAR-based object detection apparatus of claim 16 .