LANE IDENTIFICATION METHOD AND RELATED APPARATUS THEREOF

• US 20160148061A1
• Filed: 02/01/2016
• Published: 05/26/2016
• Est. Priority Date: 01/08/2013
• Status: Active Grant

• Current Assignee: PixArt Imaging Incorporated
• Original Assignee: PixArt Imaging Incorporated
• Inventors: SU, Tse-Chung
• Granted Patent: US 9,785,844 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 2207/30256 Lane; Road marking
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
• CPC Class Codes: G08G 1/167 Driving aids for lane monitoring, lane changing, e.g. blind spot detection

Abstract

A lane identification method for identifying a lane whereon a vehicle is traveling, which includes: providing a drive video; generating an identification result of recognizing a plurality of lane lines corresponding to the line according the drive video, wherein the identification result is generated based on inside borderlines of the lane lines; and determining the lane according to the identification result.

Claims

1. A lane identification method for identifying a lane whereon a vehicle is traveling, comprising:

providing a drive video;

generating an identification result of recognizing a plurality of lane lines corresponding to the line according the drive video, wherein the identification result is generated based on inside borderlines of the lane lines; and

determining the lane according to the identification result.

2. The lane identification method of claim 1, wherein the step of generating the identification result comprises:

generating a processed drive video which includes a plurality of objects.

3. The lane identification method of claim 2, wherein the step of generating the processed drive video comprises:

performing a pre-processing operation upon the drive video, which selects the objects from the drive video that meet a specific brightness requirement and a specific geometry requirement.

4. The lane identification method of claim 2, wherein the step of generating the identification result comprises:

determining a video central axis of the processed drive video;

selecting a plurality of candidate lane line segments from the objects according to the video central axis; and

generating the identification result according to the plurality of candidate lane line segments.

5. The lane identification method of claim 4, wherein the step of selecting the plurality of candidate lane line segments from the objects according to the video central axis comprises:

when there is no other object between an inside borderline of a specific object of the objects and the video central axis, determining that the specific object is one of the candidate lane line segments.

6. The lane identification method of claim 5, wherein the step of determining that the specific object is one of the candidate lane line segments comprises:

when the specific object is located at a left side of the video center axis, determining the specific object is one of candidate the lane line segments if there is no other object between a right borderline of the specific object and the video central axis; and

when the specific object is located at a right side of the video center axis, determining the specific object is one of the candidate lane line segments if there is no other object between a left borderline of the specific object and the video central axis.

7. The lane identification method of claim 4, wherein the step of generating the identification result according to the candidate lane line segments comprises:

checking if there is any specific geometric relationship between the candidate lane line segments; and

treating the candidate lane line segments as a lane line when there is specific geometric relationship between the candidate lane line segments.

8. The lane identification method of claim 7, wherein the specific geometric relationship comprises angle relationship with respect to the video central axis and neighboring relationship with respect to other candidate lane line segments.

9. A lane identification apparatus for identifying a lane whereon a vehicle is traveling, comprising:

an identification module, for generating an identification result of recognizing a plurality of lane lines corresponding to the line according a drive video, wherein the identification result is generated based on inside borderlines of the lane lines; and

a determination module, coupled to the identification module, for determining the lane according to the identification result.

10. The lane identification apparatus of claim 9, further comprising:

a video processing module, for generating a processed drive video which includes a plurality of objects.

11. The lane identification apparatus of claim 10, wherein the video processing module performs a pre-processing operation upon the drive video, which selects the objects from the drive video that meet a specific brightness requirement and a specific geometry requirement.

12. The lane identification apparatus of claim 10, wherein the identification module determines a video central axis of the processed drive video;

selects a plurality of candidate lane line segments from the objects according to the video central axis; and

generates the identification result according to the plurality of candidate lane line segments.

13. The lane identification apparatus of claim 12, wherein the identification module determines a specific object is one of the candidate lane line segments when there is no other object between an inside borderline of a specific object of the objects and the video central axis.

14. The lane identification apparatus of claim 13, wherein:

when the specific object is located at a left side of the video center axis, the identification module determines the specific object is one of candidate the lane line segments if there is no other object between a right borderline of the specific object and the video central axis; and

when the specific object is located at a right side of the video center axis, the identification module determines the specific object is one of the candidate lane line segments if there is no other object between a left borderline of the specific object and the video central axis.

15. The lane identification apparatus of claim 12, wherein the identification module checks if there is any specific geometric relationship between the candidate lane line segments;

treats the candidate lane line segments as a lane line when there is specific geometric relationship between the candidate lane line segments.

16. The lane identification apparatus of claim 15, wherein the specific geometric relationship comprises angle relationship with respect to the video central axis and neighboring relationship with respect to other candidate lane line segments.