Camera-based lane marker detection

• US 20100014714A1
• Filed: 07/18/2008
• Published: 01/21/2010
• Est. Priority Date: 07/18/2008
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations Incorporated (Motors Liquidation Co. GUC Trust)
• Inventors: Zhang, Wende, Sadekar, Varsha
• Granted Patent: US 8,204,277 B2
• Time in Patent Office: Days
• US Class Current: 382/104
• 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

Abstract

A method is provided for detecting road lane markers in a vehicle road using an imaging device. Road input data is captured using the imaging device. Lighting normalization is applied to the road input data. The method detects the road lane markers in a few main orientations in the normalized input data. In each main orientation, the normalized input data is convolved with an oriented edge detection filter for generating an oriented edge-based filter response. The normalized input data is convolved with an oriented line detection filter for generating an oriented line-based filter response. Candidate lane markers are selected in response to the edge-based filter response and line-based filter response in each main orientation. A transformation technique is applied to the candidate lane markers for identifying the lane markings in each main orientation.

Claims

1. A method of detecting road lane markers in a vehicle road using an imaging device, the method comprising the steps of:

capturing road input data using the imaging device;

applying lighting normalization to the road input data;

convolving the normalized input data with an edge detection filter for generating an edge-based filter response;

convolving the normalized input data with a line detection filter for generating a line-based filter response;

selecting candidate lane markers in response to the edge-based filter response and line-based filter response; and

applying a transformation technique to the candidate lane markers for identifying the lane markings.

2. The method of claim 1 further comprising the step of applying a false alarm mitigation analysis for verifying whether the identified lane marker is a lane marker, wherein the false alarm mitigation analysis comprises the steps of:

determining a length of the identified lane markers;

comparing the length to a predetermined length; and

determining that the identified lane markers are the lane markers in response to the length being greater than the predetermined length, otherwise determining that the identified lane markers are not the lane marker.

3. The method of claim 1 further comprising the step of applying a false alarm mitigation analysis for verifying whether the identified lane marker is a lane marker, wherein the false alarm mitigation analysis comprises the steps of:

determining a distance between a first identified lane marker and a second identified lane marker;

comparing the distance to a predetermined width, the predetermined width being representative of at least the width of a lane of a road; and

determining that the first lane marker and the second lane marker are lane markers in response to the width being less than the predetermined width, otherwise determining that the first lane marker and second lane marker are not lane markers.

4. The method of claim 3 wherein the first lane marker and the second lane marker must be parallel to one another.

5. The method of claim 1 wherein the imaging device comprises a camera.

6. The method of claim 1 wherein the processing the captured data for identifying the lane markers is performed in real time.

7. The method of claim 1 wherein the captured road input data is segmented into a plurality of straight lines oriented at different angles representing curvatures of potential lane markers in the vehicle road.

8. The method of claim 7 wherein the edge detection filter includes a plurality of oriented edge-based filters, wherein each oriented edge-based filter is applied to each segmented straight line.

9. The method of claim 8 wherein applying each oriented edge-based filter to each respective segmented straight line is performed substantially simultaneously.

10. The method of claim 7 wherein the edge detection filter outputs an enhanced edge-based filter response on each side the candidate lane marker.

11. The method of claim 10 wherein the enhanced edge-based filter response includes a first edge and a second edge of the candidate lane marker, the first edge is identified by a negative peak in the edge-based filter response and the second edge is identified by a positive peak in the edge-based filter response.

12. The method of claim 1 wherein the line detection filter includes a plurality of oriented line-based filters, wherein each oriented line-based filter is applied to each respective segmented straight line.

13. The method of claim 12 wherein the processing of the filter responses for each oriented line-based filter is performed substantially simultaneously.

14. The method of claim 12 wherein the edge-based filter response includes a positive line response, and the positive line response is identified by only a single positive peak.

15. The method of claim 1 further comprising the step of identifying a location of each of the lane markers in an output device, wherein the output device includes an image display device.

16. The method of claim 1 further comprising the step of identifying a location of each of the lane markers in an output device, wherein the output device includes an autonomous steering module.

17. A lane marker detection system comprising:

an imaging device for capturing road input data;

a processor for receiving the captured road input data received by the imaging device, the processor applying lighting normalization to the road input data, the processor convolving the normalized input data with an oriented edge detection filter for generating an edge-based filter response, the processor convolving the normalized input data with an oriented line detection filter for generating a line-based filter response, the processor applying a transformation technique to the edge-based filter response and the lane-based filter response for identifying the lane markings; and

an output device for identifying a location of each of the lane markers.

18. The lane marker detection system of claim 17 further comprising a false alarm mitigation module for verifying whether the identified lane marker resulting from the applied transformation technique is a lane marker.

19. The lane marker detection system of claim 18 wherein the imaging device is a camera.

20. The lane marker detection system of claim 17 wherein the transformation technique includes Hough transformation technique.