SYSTEM AND METHOD FOR RECOGNIZING PARKING SPACE LINE MARKINGS FOR VEHICLE

• US 20140112542A1
• Filed: 12/14/2012
• Published: 04/24/2014
• Est. Priority Date: 10/24/2012
• Status: Active Grant

• Current Assignee: Hyundai Motor Company (Hyundai Motor Group), Industry-University Cooperation Foundation Hanyang University, Kia Motors Corporation (Kia Corp.)
• Original Assignee: Hyundai Motor Company (Hyundai Motor Group), Industry-University Cooperation Foundation Hanyang University, Kia Motors Corporation (Kia Corp.)
• Inventors: Choi, Jae Seob, Yoon, Dae Joong, Chang, Eu Gene, Jung, Ho Gi
• Granted Patent: US 9,076,047 B2
• Time in Patent Office: Days
• US Class Current: 382/104
• CPC Class Codes: G06V 20/586 of parking space

Abstract

Disclosed herein is a method for recognizing a parking space line marking for a vehicle, including: detecting, by a processor, a plurality of parking spaces from a portion of a parking space line marking in an image; calculating, by the processor, an overlap coefficient representing a degree of overlapping between the detected parking spaces; selecting, by the processor, a parking space having a largest brightness coefficient as a final parking space by determining overlap when the overlap coefficient has a predetermined magnitude and comparing the brightness degrees of the overlapped parking spaces.

Claims

1. A method for recognizing a parking space line marking for a vehicle, comprising:

detecting, by a processor, a plurality of parking spaces from a portion of a parking space line marking in an image;

calculating, by the processor, an overlap coefficient representing a degree of overlap between the detected plurality of parking spaces;

calculating, by the processor, a brightness coefficient of the detected plurality of parking spaces; and

selecting, by the processor, one parking space from the detected plurality of parking spaces having a largest brightness coefficient as a final parking space by determining an overlap when the overlap coefficient includes a predetermined dimension and comparing brightness degrees of the overlapped parking spaces.

2. The method of claim 1, wherein each parking space is an end edge of the parking space line marking comprising three sides.

3. The method of claim 1, wherein the processor calculates the overlap coefficient by the following equation:

4. The method of claim 1, wherein the processor calculates the brightness coefficient by the following equation:

5. The method of claim 1, further comprising calculating, by the processor, a movement degree of the detected plurality of parking spaces in two images having a parallax error and reflecting a plurality of previous parking spaces of a previous image to a subsequent image with a plurality of subsequent parking spaces using the calculated movement degree.

6. The method of claim 5, further comprising calculating, by the processor, respective overlapping degrees between the reflected plurality of previous parking spaces and the plurality of subsequent parking spaces.

7. The method of claim 6, further comprising determining, by the processor, the overlap according to the respective overlapping degrees between the plurality of previous parking spaces and the plurality of subsequent parking spaces and determining a final parking space by adding a new parking space

8. The method of claim 6, further comprising determining, by the processor, the overlap according to the respective overlapping degrees between the plurality of previous parking spaces and the plurality of subsequent parking spaces and determining a final parking space by selecting a parking space according to the judged result.

9. The method of claim 6, further comprising adding, by the processor, a subsequent parking space as the final parking space by determining the subsequent parking space as a new parking space, when determined that the same parking space is not overlapped.

10. The method of claim 6, further comprising selecting, by the processor, the parking space having the largest brightness coefficient of a previous parking space and the subsequent parking space as the final parking space, when determined that the same parking space is overlapped.

11. The method of claim 6, further comprising selecting, by the processor, the parking space having the largest brightness coefficient of the different parking spaces as the final parking space, when determined that the different parking spaces are overlapped.

12. The method of claim 11, further comprising selecting, by the processor, a frequently selected parking space as the final parking space when determined, by the processor, that the different parking spaces are overlapped, when the parking space is continuously updated with a time interval.

13. The method of claim 7, wherein the processor detects the plurality of parking spaces by dividing a plurality of parking space types according to a plurality of shapes of the parking space line markings, and determines a final shape of the parking space line marking according to the shape corresponding to the selected final parking space.

14. A controller for recognizing a parking space line marking for a vehicle, comprising:

a memory; and

a processor configured to;

detect a plurality of parking spaces from a portion of a parking space line marking in an image;

calculate an overlap coefficient representing a degree of overlap between the detected parking spaces;

calculate a brightness coefficient of the detected parking spaces; and

select a parking space from the detected parking spaces having a largest brightness coefficient as a final parking space by determining an overlap when the overlap coefficient includes a predetermined dimension and comparing brightness degrees of the overlapped parking spaces.

15. The controller of claim 14 wherein the processor is further configured to:

calculate the overlap coefficient using a sum of lengths of an i-th parking space and a j-th parking space and a length of an overlapped portion of the i-th parking space and the j-th parking space;

calculate the brightness coefficient using a brightness value of a plurality of points of the i-th parking space and an l-th parking space and determine a largest brightness value;

calculate a movement degree of the detected plurality of parking spaces in two images having a parallax error and reflecting a plurality of previous parking spaces of a previous image to a subsequent image with a plurality of subsequent parking spaces using the calculated movement degree; and

calculate respective overlapping degrees between the reflected plurality of previous parking spaces and the plurality of subsequent parking spaces.

16. The controller of claim 14, wherein the processor is further configured to detect the plurality of parking spaces by dividing a plurality of parking space types according to a plurality of shapes of the parking space line marking, and determine a final shape of the parking space line marking according to the shape corresponding to the selected final parking space.

17. A non-transitory computer readable medium containing program instructions executed by a processor or controller, the computer readable medium comprising:

program instructions that detect a plurality of parking spaces from a portion of a parking space line marking in an image;

program instructions that calculate an overlap coefficient representing a degree of overlap between the detected parking spaces;

program instructions calculate a brightness coefficient of the detected parking spaces; and

program instructions select a parking space from the detected parking spaces having a largest brightness coefficient as a final parking space by determining an overlap when the overlap coefficient includes a predetermined dimension and comparing brightness degrees of the overlapped parking spaces.

18. The non-transitory computer readable medium of claim 17, further comprising:

program instructions that calculate the overlap coefficient using a sum of lengths of an i-th parking space and a j-th parking space and a length of an overlapped portion of the i-th parking space and the j-th parking space;

program instructions that calculate the brightness coefficient using a brightness value of a plurality of points of the i-th parking space and an l-th parking space and determine a largest brightness value;

program instructions that calculate a movement degree of the detected plurality of parking spaces in two images having a parallax error and reflecting a plurality of previous parking spaces of a previous image to a subsequent image with a plurality of subsequent parking spaces using the calculated movement degree; and

program instructions that calculate respective overlapping degrees between the reflected plurality of previous parking spaces and the plurality of subsequent parking spaces.

19. The non-transitory computer readable medium of claim 17, further comprising program instructions that detect the plurality of parking spaces by dividing a plurality of parking space types according to a plurality of shapes of the parking space line marking, and determine a final shape of the parking space line marking according to the shape corresponding to the selected final parking space.