Detection of obstacles at night by analysis of shadows

• US 9,297,641 B2
• Filed: 12/11/2012
• Issued: 03/29/2016
• Est. Priority Date: 12/12/2011
• Status: Expired

• Current Assignee: MobilEye Vision Technologies Ltd. (Intel Corporation)
• Original Assignee: MobilEye Vision Technologies Ltd. (Intel Corporation)
• Inventors: Stein, Gideon
• Primary Examiner(s): Tran, Thai
• Assistant Examiner(s): MESA, JOSE M
• Application Number:
  US13/710,564
• Publication Number:
  US 20130147957A1
• Time in Patent Office: 1,204 Days
• Field of Search: 348/148
• US Class Current: 1/1
• CPC Class Codes: G01B 11/02 for measuring length, width or thickness G01B11/08 takes precedence
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads

Abstract

A driver assistance systems mountable in a host vehicle while the host vehicle is moving forward with headlights on for detection of obstacles based on shadows. The driver assistance system includes a camera operatively connectible to a processor. A first image frame and a second image frame are captured of a road. A first dark image patch and a second dark image patch include intensity values less than a threshold value. The first and the second dark image patches are tracked from the first image frame to the second image frame as corresponding images of the same portion of the road. Respective thicknesses in vertical image coordinates are measured of the first and second dark image patches responsive to the tracking.

Claims

1. A computerized method performable at night by a driver assistance system mountable in a host vehicle while the host vehicle is moving forward with headlights on, wherein the driver assistance system includes a camera operatively connectible to a processor, the method comprising:

capturing a first image frame and a second image frame of a road in the field of view of the camera;

processing said first image frame and said second image frame to locate a first dark image patch of said first image frame and a second dark image patch of said second image frame;

tracking said first and said second dark image patches from said first image frame to said second image frame as corresponding images of the same portion of the road;

measuring respective thicknesses in vertical image coordinates of said first and second dark image patches responsive to said tracking from said first image frame to said second image frame; and

determining that a change of said thicknesses between said first and second dark image patches is consistent with a shadow cast by an obstacle in the road illuminated by the headlights.

2. The method of claim 1, wherein said processing is performed by filtering said first image frame and said second image frame with a threshold value of gray scale to produce said first dark image patch of said first image frame and said second dark image patch of said second image frame, wherein said first dark image patch and said second dark image patch include intensity values less than said threshold value.

3. The method of claim 1, further comprising:

prior to said tracking, verifying connectivity of picture elements within said first dark image patch.

4. The method of claim 1, further comprising:

determining respective distances Z from the camera to the obstacle during the capturing of the first and the second image frames, wherein the thicknesses change proportionally to inverse distance Z<sup>−

1</sup> or the thicknesses change less than proportionally to inverse distance Z<sup>−

1</sup> thereby detecting that the first and second dark image patches are the shadow cast by the obstacle in the road illuminated by the headlights.

5. The method of claim 4, further comprising:

audibly warning a driver of the host vehicle responsive to said detecting said obstacle.

6. The method of claim 4, further comprising:

analyzing change in said thicknesses between said first and second image frames, thereby distinguishing between the detected obstacle being a positive or a negative obstacle.

7. The method of claim 4, further comprising:

analyzing change in said thicknesses between said first and second image frames, thereby computing a shape of the detected obstacle.

8. The method of claim 1, further comprising:

determining that a change of said thicknesses between said first and second image frames is not consistent with a shadow cast by an obstacle illuminated by the headlights.

9. The method of claim 1, wherein said thicknesses are measured in picture elements according to a total number of picture elements with intensity values less than said threshold divided by a width in picture element columns of said first and second dark image patches.

10. A driver assistance system including a camera and a processor configured to perform the computerized method, according to claim 1.

16. The method of claim 8, further comprising determining that a change of said thicknesses between said first and second dark image patches is consistent with a marking in the road by determining respective distances from the camera to the marking in the road during the capturing of the first and the second image frames, wherein the thicknesses change proportionally to inverse square distance to the marking and detecting thereby that the first and second dark image patches are images of the marking in the road.

11. A driver assistance system mountable in a host vehicle, the driver assistance system comprising:

a processor;

a camera operatively connectible to the processor, while the host vehicle is moving with headlights on, the processor is configured to capture from the camera a first image frame and a second image frame of a road in the field of view of the camera;

wherein the processor is configured to process said first image frame and said second image frame to locate a first dark image patch of said first image frame and a second dark image patch of said second image frame;

wherein the processor is configured to track said first and second dark image patches from said first image frame to said second image portion of said second image frame as corresponding images of the same portion of the road;

wherein the processor is configured to measure respective thicknesses in vertical image coordinates of said first and second dark image patches responsive to said tracking from said first image frame to said second image frame; and

wherein the processor is configured to determine that a change of said thicknesses between said first and second dark image patches is consistent with a shadow cast by an obstacle in the road illuminated by the headlights to detect the obstacle in the road.

12. The driver assistance system of claim 11, wherein respective distances Z from the camera to the obstacle are determinable during the capture of the first and the second image frames, wherein the thicknesses change proportionally to inverse distance Z⁻

1 or the thicknesses change less than proportionally to inverse distance Z<sup>−

1</sup> to detect that the first and second dark image patches are the shadow cast by the obstacle in the road illuminated by the headlights.

13. The driver assistance system of claim 12, wherein the processor is configured to analyze change in said thicknesses between said first and second image frames to distinguish between the detected obstacle being a positive or a negative obstacle.

14. The driver assistance system of claim 12, wherein the processor is configured to analyze change in said thicknesses between said first and second image frames to compute a shape of the detected obstacle.

15. The driver assistance system of claim 11, wherein the processor is configured to determine that a change of said thicknesses between said first and second image frames is not consistent with a shadow cast by an obstacle illuminated by the headlights.

17. The driver assistance system of claim 15, wherein respective distances from the camera to a marking in the road are determinable during the capture of the first and the second image frames, wherein the thicknesses change proportionally to inverse square distance to the marking to detect that the first and second dark image patches are images of the marking in the road.