IP Library › Granted Patent US 11,321,950
Granted Patent B2
US 11,321,950 · App. 17/057,498 · Granted May 3, 2022

Apparatus and method for detecting lane information, and computer-readable recording medium storing computer program programmed to execute same method

Inventor: Seong Soo Lee (Seoul, KR)
Assignee: SK TELECOM CO., LTD.
G06V20/588B60W30/12B60W40/04B60W40/076B60W2420/42
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,321,950
App. No.
17/057,498
Granted
May 3, 2022
Kind
B2
Abstract

In accordance with an aspect of the present disclosure, there is provided a method for detecting a lane information comprising: obtaining, from a high definition map, first driving lane information corresponding to estimated position information on a vehicle; obtaining second driving lane information from a front view image from the vehicle captured by a camera installed in the vehicle; converting the first driving lane information and the second driving lane information according to an identical coordinate system; and obtaining final driving lane information by combining the converted first driving lane information and the converted second driving lane information.

Claims (70)

1. A lane information detecting method comprising:

obtaining, from a high definition map, first driving lane information corresponding to estimated position information on a vehicle;

obtaining second driving lane information from a front view image from the vehicle captured by a camera installed in the vehicle;

converting the first driving lane information and the second driving lane information according to an identical coordinate system; and

obtaining final driving lane information by combining the converted first driving lane information and the converted second driving lane information.

2. The method of claim 1 , wherein obtaining the first driving lane information from the high definition map includes:

obtaining estimated position information on the camera based on a matching error between a landmark in the high definition map corresponding to each of a plurality of candidate position information sampled based on an initial position of the camera and the front view image;

obtaining the estimated position information on the vehicle based on estimated position information on the camera;

determining a driving lane of the vehicle based on the estimated position information on the vehicle; and

obtaining, from the high definition map, the first driving lane information on both sides of the determined driving lane.

3. The method of claim 1 , wherein the first driving lane information is obtained based on a reference coordinate system, and the second driving lane information is obtained based on an image coordinate system for the front view image.

4. The method of claim 1 , wherein converting the first driving lane information and the second driving lane information according to the identical coordinate system is converting the first driving lane information and the second driving lane information according to a vehicle coordinate system by using the estimated position information on the vehicle.

5. The method of claim 4 , wherein converting the second driving lane information according to the vehicle coordinate system includes:

obtaining the second driving lane information in a top view image corresponding to the front view image based on the estimated position information on the vehicle; and

converting the second driving lane information in the top view image according to the vehicle coordinate system.

6. The method of claim 4 , wherein converting the second driving lane information according to the vehicle coordinate system includes:

gridding surroundings of estimated position information on the camera on the high definition map into a plurality of planes;

obtaining a vector directed to the second driving lane information in the front view image from an origin of a camera coordinate system by using the estimated position information on the vehicle;

obtaining, based on a straight line determined from the obtained vector and the plurality of the gridded planes, the second driving lane information according to a reference coordinate system; and

converting, based on the estimated position information on the vehicle, the second driving lane information according to the reference coordinate system into the second driving lane information according to the vehicle coordinate system.

7. The method of claim 6 ,

wherein a plane is classified into a first plane group when position information of four or more points within one plane are obtainable, and is classified into a second plane group when the position information of three or less points within the one plane are obtainable, and

wherein gridding the surroundings of the estimated position information on the camera on the high definition map into the plurality of the planes includes:

determining a plane equation for a first plane in the first plane group based on position information of four or more points on the first plane; and

determining a plane equation for a second plane in the second plane group based on a plane equation of a plane which belongs to the first group and adjacent to the second plane.

8. The method of claim 4 , wherein obtaining the final driving lane information includes:

obtaining combined driving lane information by combining the converted first driving lane information and the converted second driving lane information;

performing sampling for the obtained combined driving lane information; and

obtaining the final driving lane information by performing fitting for the sampled combined driving lane information with a polynomial function.

9. The method of claim 8 , wherein obtaining the combined driving lane information includes:

identifying an overlapping area of the converted first driving lane information and the converted second driving lane information; and

obtaining the combined driving lane information based on the estimated position information on the vehicle by selecting, in the overlapping area, one of the converted first driving lane information and the converted second driving lane information.

10. The method of claim 1 , wherein converting the first driving lane information and the second driving lane information according to the identical coordinate system is converting the first driving lane information according to an image coordinate system for the front view image.

11. The method of claim 10 , wherein obtaining the final driving lane information includes:

mapping the first driving lane information converted according to the image coordinate system on the front view image;

identifying an overlapping area of the first driving lane information and the second driving lane information in the front view image;

obtaining combined driving lane information by selecting, for the overlapping area, one of the first driving lane information and the second driving lane information based on a reference distance; and

obtaining the final driving lane information by performing fitting for the combined driving lane information.

12. A non-transitory computer-readable recording medium storing instructions, when executed by a processor to cause the processor to perform a lane information detecting method, the method comprising:

obtaining, from a high definition map, first driving lane information corresponding to estimated position information on a vehicle;

obtaining second driving lane information from a front view image from the vehicle captured by a camera mounted in the vehicle;

converting the first driving lane information and the second driving lane information according to an identical coordinate system; and

obtaining final driving lane information by combining the converted first driving lane information and the converted second driving lane information.

13. The recording medium of claim 12 , wherein obtaining the first driving lane information from the high definition map includes:

obtaining estimated position information on the camera based on a matching error between a landmark in the high definition map corresponding to each of a plurality of candidate position information sampled based on an initial position of the camera and the front view image;

obtaining the estimated position information on the vehicle based on estimated position information on the camera;

determining a driving lane of the vehicle based on the estimated position information on the vehicle; and

obtaining, from the high definition map, the first driving lane information on both sides of the determined driving lane.

14. The recording medium of claim 12 , wherein the first driving lane information is obtained based on a reference coordinate system, and the second driving lane information is obtained based on an image coordinate system for the front view image.

15. The recording medium of claim 12 , wherein converting the first driving lane information and the second driving lane information according to the identical coordinate system is converting the first driving lane information and the second driving lane information according to a vehicle coordinate system by using the estimated position information on the vehicle.

16. The recording medium of claim 15 , wherein converting the second driving lane information according to the vehicle coordinate system includes:

obtaining the second driving lane information in a top view image corresponding to the front view image based on the estimated position information on the vehicle; and

converting the second driving lane information in the top view image according to the vehicle coordinate system.

17. The recording medium of claim 15 , wherein converting the second driving lane information according to the vehicle coordinate system includes:

gridding surroundings of estimated position information on the camera on the high definition map into a plurality of planes;

obtaining a vector directed to the second driving lane information in the front view image from an origin of a camera coordinate system by using the estimated position information on the vehicle;

obtaining, based on a straight line determined from the obtained vector and the plurality of the gridded planes, the second driving lane information according to a reference coordinate system; and

converting, based on the estimated position information on the vehicle, the second driving lane information according to the reference coordinate system into the second driving lane information according to the vehicle coordinate system.

18. The recording medium of claim 17 , wherein a plane is classified into a first plane group when position information of four or more points within one plane are obtainable, and is classified into a second plane group when the position information of three or less points within the one plane are obtainable, and

wherein gridding the surroundings of the estimated position information on the camera on the high definition map into the plurality of the planes includes:

determining a plane equation for a first plane in the first plane group based on position information of four or more points on the first plane; and

determining a plane equation for a second plane in the second plane group based on a plane equation of a plane which belongs to the first group and adjacent to the second plane.

19. The recording medium of claim 15 , wherein obtaining the final driving lane information includes:

obtaining combined driving lane information by combining the converted first driving lane information and the converted second driving lane information;

performing sampling for the obtained combined driving lane information; and

obtaining the final driving lane information by performing fitting for the sampled combined driving lane information with a polynomial function.

20. A lane information detecting apparatus comprising:

a driving lane information obtaining unit configured to obtain, from a high definition map, first driving lane information corresponding to estimated position information on a vehicle, and obtain second driving lane information from a front view image from the vehicle captured by a camera installed in the vehicle;

a coordinate system converting unit configured to convert the first driving lane information and the second driving lane information according to an identical coordinate system; and

a lane combining unit configured to obtain final driving lane information by combining the converted first driving lane information and the converted second driving lane information.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 20, 2020
From: LEE, SEONGSOO
To: SK TELECOM CO., LTD.
Reel/Frame 054492/0734 →
Priority Claims (1)
KR 10-2018-0073259 · Jun 26, 2018 · national
Continuity (1)
Related Publication 20210365694A1 · Nov 25, 2021
Cited By (2)
US 12,233,876 US 12,511,717