Method and apparatus for removing LiDAR point cloud noise
Disclosed is a method and apparatus for removing LiDAR point cloud noise. This method removes in real time only the LiDAR point cloud noise acquired in snowfall and rain environments, leaving behind the environmental characteristics of the LiDAR point cloud. Accordingly, it is possible to maximize the utilization of the LiDAR.
1. A method for removing LiDAR point cloud noise, comprising the steps of:
(a) checking whether a distance of a LiDAR point is within a noise detection distance;
(b) if the distance of the LiDAR point is within the noise detection distance in step (a), checking whether an intensity measurement value of the LiDAR point is less than a noise intensity threshold;
(c) if there is a specific LiDAR point whose intensity measurement value is less than the noise intensity threshold in step (b), checking whether a number of neighboring LiDAR points within an arbitrary radius of the specific LiDAR point is greater than or equal to a certain threshold value; and
(d) if the number of neighboring LiDAR points within the arbitrary radius of the specific LiDAR point is not greater than or equal to the certain threshold value in step (c), determining the specific LiDAR point as noise to be removed,
wherein the noise intensity threshold is a value obtained through a reflectance and incident angle of noise, and an area ratio between a footprint size of a LiDAR beam and a noise size.
2. A method according to claim 1 ,
wherein, if the distance of the LiDAR point is not within the noise detection distance in step (a), maintaining the LiDAR point as environmental information.
3. The method according to claim 1 ,
wherein, if the intensity measurement value of the LiDAR point is greater than the noise intensity threshold in step (b), maintaining the LiDAR point as environmental information.
4. The method according to claim 1 ,
wherein the noise detection distance is 71.235 m.
5. The method according to claim 1 ,
wherein the arbitrary radius is 0.1 m.
6. The method according to claim 1 ,
wherein the certain threshold value is 3 pts.
7. The method according to claim 1 ,
wherein, in step (c), classifying the specific LiDAR point as noise before checking whether the number of neighboring LiDAR points within the arbitrary radius of the specific LiDAR point is greater than or equal to the certain threshold value.
8. The method of claim 7 ,
wherein, in step (d), if the number of neighboring LiDAR points within the arbitrary radius of the specific LiDAR point is greater than or equal to the certain threshold value as the result of the checking in step (c), reclassifying the specific LiDAR point classified as noise before as environmental information.
9. An apparatus for removing LiDAR point cloud noise, comprising:
at least one processor; and,
at least one memory storing computer-executable instructions,
wherein the computer-executable instructions stored in said at least one memory, when executed by the at least one processor, causes the at least one processor to perform operations comprising the steps of:
(a) checking whether a distance of a LiDAR point is within a noise detection distance;
(b) if the distance of the LiDAR point is within the noise detection distance in step (a), checking whether an intensity measurement value of the LiDAR point is less than a noise intensity threshold;
(c) if there is a specific LiDAR point whose intensity measurement value is less than the noise intensity threshold in step (b), checking whether a number of neighboring LiDAR points within an arbitrary radius of the specific LiDAR point is greater than or equal to a certain threshold value; and
(d) if the number of neighboring LiDAR points within the arbitrary radius of the specific LiDAR point is not greater than or equal to the certain threshold value in step (c), determining the specific LiDAR point as noise to be removed,
wherein the noise intensity threshold is a value obtained through a reflectance and incident angle of noise, and an area ratio between a footprint size of a LiDAR beam and a noise size.