Barrier and guardrail detection using a single camera
Driver assistance systems for detecting a structural barrier extending along a road. The driver assistance system may be mountable in a host vehicle. The camera may capture multiple image frames in the forward field of view of the camera. A processor may process motion of images of the barrier in the image frames. The camera may be a single camera. The motion of the images may be responsive to forward motion of the host vehicle and/or the motion of the images may be responsive to lateral motion of the host vehicle.
1. An image processing method, comprising:
obtaining, via a camera located in a moving vehicle, a first image frame;
locating, in the first image frame, a first image patch comprising an image line intersecting a vanishing point in the first image frame;
obtaining, via the camera located in the moving vehicle, a second image frame;
locating, in the second image frame, a second image patch corresponding to the first image patch;
warping the first image patch based on a vertical surface model and known vehicle motion to provide a first warped image patch; and
identifying a match between the second image patch and the first warped image patch.
2. The method according to claim 1 , wherein identifying a match between the second image patch and the first warped image patch comprises:
identifying, in the first image frame or in the second frame image, image data corresponding to a vertical structure.
3. The method according to claim 2 , further comprising:
determining a lateral distance between the vehicle and the vertical structure.
4. The method according to claim 1 , further comprising:
warping the first image patch based on a horizontal surface model to provide a second warped image patch, and
wherein identifying a match between the second image patch and the first warped image patch comprises:
identifying, in the first image frame or in the second frame image, image data corresponding to a horizontal structure.
5. The method according to claim 4 , wherein identifying a match between the second image patch and the first warped image patch comprises:
determining a matching score of the vertical surface model and the image data, and
determining a matching score of the horizontal surface model and the image data.
6. The method according to claim 4 , further comprising:
determining a longitudinal distance between the vehicle and the vertical structure.
7. The method according to claim 1 , wherein the first image patch is on or bounded by an image line intersecting a vanishing point in the first image frame, and the second image patch is on or bounded by an image line intersecting a vanishing point in the second image frame.
8. The method according to claim 7 , wherein the image line intersecting the vanishing point in the first image frame is associated with a barrier lacking substantial vertical image texture, and the image line intersecting the vanishing point in the second image frame is associated with a barrier lacking substantial vertical image texture.
9. An image processing system, comprising:
a camera configured to be placed in a moving vehicle and obtain a first image frame; and
a processor operatively connected to the camera, the processor being operable to:
locate, in the first image frame, a first image patch comprising an image line intersecting a vanishing point in the first image frame;
obtain, via the camera located in the moving vehicle, a second image frame;
locate, in the second image frame, a second image patch corresponding to the first image patch in the first image frame;
warp the first image patch based on a vertical surface model and known vehicle motion to provide a first warped image patch; and
identify a match between the second image patch and the first warped image patch.
10. The image processing system according to claim 9 , wherein the processor is operable to identify, in the first image frame or in the second frame image, image data corresponding to a vertical structure.
11. The image processing system according to claim 10 , wherein the processor is operable to determine a lateral distance between the vehicle and the vertical structure.
12. The image processing system according to claim 9 , wherein the processor is operable to warp the first image patch based on a horizontal surface model to provide a second warped image patch, and to identify, in the first image frame or in the second frame image, image data corresponding to a horizontal structure.
13. The image processing system according to claim 12 , wherein the processor is operable to determine a matching score of the vertical surface model and the image data, and to determine a matching score of the horizontal surface model and the image data.
14. The image processing system according to claim 12 , wherein the processor is operable to determine a longitudinal distance between the vehicle and the vertical structure.
15. The image processing system according to claim 9 , wherein the first image patch is on or bounded by an image line intersecting a vanishing point in the first image frame, and the second image patch is on or bounded by an image line intersecting a vanishing point in the second image frame.
16. The image processing system according to claim 15 , wherein the image line intersecting the vanishing point in the first image frame is associated with a barrier lacking substantial vertical image texture, and the image line intersecting the vanishing point in the second image frame is associated with a barrier lacking substantial vertical image texture.