Systems and methods for curb detection and pedestrian hazard assessment

Patent metadata

Field	Value
Publication number	US-9443163-B2
Application number	US-201514712580-A
Country	US
Kind code	B2
Filing date	May 14, 2015
Priority date	May 14, 2014
Publication date	Sep 13, 2016
Grant date	Sep 13, 2016

Abstract

A detection system for a vehicle is provided. The detection system may include at least one image capture device configured to acquire a plurality of images of an area forward of the vehicle, the area including a curb separating a road surface from an off-road surface and a data interface. The detection system may also include at least one processing device programmed to receive the plurality of images via the data interface, and determine a plurality of curb edge line candidates in the plurality of images. The at least one processing device may be further programmed to identify at least one edge line candidate as an edge line of the curb.

First claim

What is claimed is:

1. A detection system for a vehicle, the system comprising:
at least one camera configured to acquire a plurality of images of an area forward of the vehicle, the area including a curb separating a road surface from an off-road surface;
a data interface; and
at least one processing device programmed to:
receive the plurality of images output from the at least one camera via the data interface;
determine a plurality of curb edge line candidates from and within the plurality of images of the area forward of the vehicle; and
identify at least one curb edge line candidate as an edge line of the curb.

2. The detection system of claim 1 , wherein the edge line of the curb is a curb bottom edge or a curb top edge.

3. The detection system of claim 1 , wherein the at least one processing device is further programmed to determine a height of a step function associated with each curb edge line candidate.

4. The detection system of claim 3 , wherein identifying the at least one curb edge line candidate as an edge line of the curb includes comparing the height of the step function to a threshold value.

5. The detection system of claim 3 , wherein determining the height includes executing an image warp function between pixel coordinates of the plurality of images.

6. The detection system of claim 1 , wherein the at least one processing device is further programmed to identify regions in the plurality of images that correspond to the road surface and the off-road surface based on the identified edge line of the curb.

7. The detection system of claim 6 , wherein the at least one processing device is further programmed to:
identify a pedestrian in one or more of the plurality of images;
assess whether the pedestrian presents a hazard; and
perform an action to avoid a collision with the pedestrian, based on the assessment.

8. The detection system of claim 7 , wherein assessing whether the pedestrian presents a hazard includes classifying the pedestrian based on the identified regions and at least one of a location or a trajectory of the pedestrian.

9. The detection system of claim 8 , wherein classifying the pedestrian includes classifying the pedestrian as on the road surface or on the off-road surface.

10. The detection system of claim 7 , wherein assessing whether the pedestrian presents a hazard includes comparing a trajectory of the pedestrian to a trajectory of a vehicle.

11. The detection system of claim 7 , wherein the action to avoid a collision includes alerting a driver.

12. The detection system of claim 7 , wherein the action to avoid the collision includes modifying at least one of a speed, acceleration, or direction of a vehicle.

13. A vehicle, comprising:
a body;
at least one camera configured to acquire a plurality of images of an area forward of the vehicle, the area including a curb separating a road surface from an off-road surface;
a data interface; and
at least one processing device programmed to:
receive the plurality of images output from the at least one camera via the data interface;
determine a plurality of curb edge line candidates from and within the plurality of images of the area forward of the vehicle;
identify at least one curb edge line candidate as an edge line of the curb; and
identify regions in the plurality of images that correspond to the road surface and the off-road surface based on the identified edge line of the curb.

14. The vehicle of claim 13 , wherein the at least one processing device is further programmed to:
identify a pedestrian in one or more of the plurality of images;
assess whether the pedestrian presents a hazard; and
perform an action to avoid a collision with the pedestrian, based on the assessment.

15. The vehicle of claim 14 , wherein assessing whether the pedestrian presents a hazard includes classifying the pedestrian based on the identified regions and a location of the pedestrian.

16. The vehicle of claim 15 , wherein classifying the pedestrian includes comparing a location of the pedestrian to the identified regions.

17. The vehicle of claim 16 , wherein classifying the pedestrian includes classifying the pedestrian as on the road surface or on the off-road surface based on the comparison.

18. The vehicle of claim 13 , wherein assessing whether the pedestrian presents a hazard includes comparing a trajectory of the pedestrian to a trajectory of a vehicle.

19. A method for curb detection, the method comprising:
acquiring, via at least one camera, a plurality of images of an area forward of a vehicle, the area including a curb separating a road surface from an off-road surface;
determining a plurality of curb edge line candidates from and within the plurality of images of the area forward of the vehicle;
determining a height of a step function associated with each curb edge line candidate; and
identifying at least one curb edge line candidate as an edge line of the curb based on the determined heights.

20. The method of claim 19 , wherein determining the height of the step function includes executing an image warp function between pixel coordinates of the plurality of images.

Assignees

• Springer Ofer
• Mobileye Vision Technologies Ltd

Inventors

• Springer Ofer

Classifications

• G08G1/166 Primary

  for active traffic, e.g. moving vehicles, pedestrians, bikes · CPC title

• G06V20/58

  Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads · CPC title

• G06F18/24

  Classification techniques · CPC title

• H04N7/183

  for receiving images from a single remote source · CPC title

• G06T2207/30261

  Obstacle · CPC title