System and method for periodic lane marker identification and tracking

• US 9,836,657 B2
• Filed: 06/25/2015
• Issued: 12/05/2017
• Est. Priority Date: 07/27/2011
• Status: Active Grant

• Current Assignee: HL Klemove Corporation
• Original Assignee: Gentex Corporation
• Inventors: Hilldore, Benjamin B., Rycenga, Brock R., TerHaar, Micah W.
• Primary Examiner(s): Bella, Matthew
• Assistant Examiner(s): TORRES, JOSE
• Application Number:
  US14/750,234
• Publication Number:
  US 20150294165A1
• Time in Patent Office: 894 Days
• Field of Search: None
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle
• 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 system and method are disclosed for determining the presence and period of dashed line lane markers in a roadway. The system includes an imager configured to capture a plurality of high dynamic range images exterior of the vehicle and a processor, in communication with the at least one imager such that the processor is configured to process at least one high dynamic range image. The period of the dashed lane markers in the image is calculated for detecting the presence of the dashed lane marker and for tracking the vehicle within the markers. The processor communicates an output for use by the vehicle for use in lane departure warning (LDW) and/or other driver assist features.

Claims

1. A system for periodic lane marker identification comprising:

at least one imager configured to capture image data corresponding to a plurality of images exterior of a vehicle;

a processor in communication with the at least one imager wherein the processor is configured to;

convert the image data from a relative coordinate system to a world coordinate system;

process a first sample of the image data in a first direction and a second sample of the image data in a second direction to identify a potential raised road marker;

define a box encompassing at least a portion of the potential raised road marker based on the first sample and the second sample; and

identify a detected raised road marker from the potential raised road marker based on an intensity of a plurality of points located in the box.

2. The system according to claim 1, wherein the first sample corresponds to a substantially horizontal slice of the image data along the first direction.

3. The system according to claim 1, wherein the second sample corresponds to a substantially vertical slice of the image data along the second direction.

4. The system according to claim 1, wherein the potential raised road marker corresponds to a dark circle of road around the raised road marker in the image data.

5. The system according to claim 1, wherein the processing of the first sample and the second sample comprises identifying a first width along the first direction and a second width along the second direction.

6. The system according to claim 5, further comprising comparing the second width to the first width to determine if the second width is sufficiently long to correspond to a potential raised road marker relative to the second width.

7. The system according to claim 1, wherein the processor continues to define the box in response to a positive identification of the potential raised road marker based on the first sample and the second sample.

8. The system according to claim 7, wherein the detected raised road marker is further identified based on a raised lane marker confidence value calculated based on the intensity of the plurality of points located in the box.

9. A system for periodic lane marker identification comprising:

at least one imager configured to capture image data corresponding to a plurality of images exterior of a vehicle;

a processor in communication with the at least one imager wherein the processor is operable to;

convert the image data from a relative coordinate system to a world coordinate system;

process a first slice of the image data in a first direction to identify a potential raised road marker;

in response to identifying the potential raised road marker in the first direction corresponding to the potential raised road marker, process a second slice in a second direction to identify a second intensity of pixels proximate to the potential raised road marker;

compare a first width of the first intensity of pixels to a second width of the second intensity of pixels to further identify the potential raised road marker; and

in response to the further identification of the raised road marker, identify a detected raised road marker from the potential raised road marker based on an intensity of a plurality of points located in a box defined by the first intensity of pixels and the second intensity of pixels.

10. The system according to claim 9, wherein the second slice is located proximate to the potential raised road marker.

11. The system according to claim 9, wherein the first direction is substantially perpendicular to the second direction.

12. The system according to claim 9, wherein the processor is further operable to:

in response to identifying that the first intensity of pixels in the first direction does not correspond to the potential raised road marker, search a different portion of the image data in the first direction to identify the first intensity of pixels corresponding to the potential raised road marker.

13. The system according to claim 9, wherein the first intensity of pixels corresponds to a horizontal width of the potential raised lane marker and the second intensity of pixels corresponds to a vertical width of the potential raised lane marker.

14. The system according to claim 13, wherein the further identification of the potential raised lane marker comprises determining if the second width is sufficiently long relative to the first width to correspond to a potential raised road marker.

15. The system according to claim 9, wherein the processor is further operable to:

identify a detected raised road marker based on a raised lane marker confidence value based on an intensity of a plurality of points of the potential raised road marker.

16. A system for periodic lane marker identification comprising:

at least one imager configured to capture image data corresponding to a plurality of images exterior of a vehicle;

a processor in communication with the at least one imager wherein the processor is configured to;

convert the image data from a relative coordinate system to a world coordinate system;

process a first slice of the image data in a horizontal direction to determine a first width;

detect a potential raised lane marker from the first slice of the image data, wherein;

upon not detecting the potential raised lane marker, the processor processes the first slice of a different portion of the image data to detect the potential raised lane marker; and

upon detecting the potential raised lane marker, the processor processes a second slice of the image data in a vertical direction to determine a second width to identify a potential raised road marker; and

further identify the potential raised road marker by comparing the first width to the second width.

17. The system according to claim 16, wherein the further identification of the potential raised lane marker comprises determining if the second width is sufficiently long relative to the first width to correspond to a potential raised road marker.

18. The system according to claim 16, wherein the processor is further configured to define a box encompassing at least a portion of the potential raised road marker based on the first width and the second width.

19. The system according to claim 18, wherein the processor is further configured to identify a detected raised road marker from the potential raised road marker based on an intensity of a plurality of points located in the box.

20. The system according to claim 19, wherein the detected raised road marker is further identified based on a raised lane marker confidence value calculated based on the intensity of the plurality of points located in the box.