Wet road surface condition detection

Patent metadata

Field	Value
Publication number	US-9972206-B2
Application number	US-201514957943-A
Country	US
Kind code	B2
Filing date	Dec 3, 2015
Priority date	Dec 3, 2015
Publication date	May 15, 2018
Grant date	May 15, 2018

Abstract

A method for determining a wet surface condition of a road. Capturing an image of a wheel of a remote vehicle traveling in an adjacent lane by an image capture device of a host vehicle. Identifying in the captured image, by processor of a host vehicle, a region of interest relative to the wheel where the region of interest is representative of where precipitation dispersion occurs. A determination is made whether precipitation is present in the region of interest. A wet road surface signal is generated in response to the identification of precipitation in the adjacent lane.

First claim

What is claimed is:

1. A method for determining a wet surface condition of a road with adjacent first and second lanes, the method comprising:
capturing an image of a road wheel of a remote vehicle traveling in the first lane by an image capture device of a host vehicle traveling in the second lane;
identifying, in the captured image by a processor of the host vehicle, a region of interest relative to the road wheel of the remote vehicle, wherein the region of interest has a location, a width, and a height, the width and height determined as a function of a tire radius of a tire of the road wheel, and the location determined as a function of the tire radius and a rim radius of a rim of the road wheel, wherein the tire radius and the rim radius are determined by:
determining a location of the road wheel in the captured image;
applying an edge detection technique to the road wheel within the captured image;
applying an image transform analysis to identify respective positions of one or more shapes associated with the road wheel in the captured image;
identifying contours of the tire and the rim in the captured image;
identifying a center of the road wheel; and
determining the tire radius and the rim radius each as a function of a respective distance from the center of the road wheel to the identified contours of the tire or the rim;

determining whether or not precipitation is present in the region of interest; and
generating a wet road surface signal in response to a determination that precipitation is present in the region of interest.

2. The method of claim 1 , wherein the region of interest is located in a region where precipitation is dispersed by the road wheel of the remote vehicle traveling in the first lane.

3. The method of claim 1 , wherein the region of interest is a rectangular region, the width is substantially equal to the tire radius, and the height is substantially equal to one half of the tire radius.

4. The method of claim 1 , wherein the region of interest extends in a direction rearward of the center of the road wheel, wherein a corner of the region of interest is located at a coordinate relative to the center of the road wheel, the coordinate being located a lateral length from the center of the road wheel substantially equal to the rim radius of the rim of the road wheel and a longitudinal length from the center of the road wheel substantially equal to the tire radius of the tire.

5. The method of claim 1 , wherein the edge detection technique detects horizontal, vertical, and diagonal edges of the road wheel within the captured image.

6. The method of claim 1 , wherein the image transform analysis includes a Hough transformation analysis applied to the edge-detected image to identify lines and positions of shapes of the road wheel.

7. A method for determining a wet surface condition of a road with adjacent first and second lanes, the method comprising:
capturing an image of a road wheel of a remote vehicle traveling in the first lane by an image capture device of a host vehicle traveling in the second lane;
identifying, in the captured image by a processor of the host vehicle, a region of interest relative to the road wheel of the remote vehicle, the region of interest having a location determined as a function of a location of the road wheel, wherein determining the location of the road wheel in the captured image includes applying a wheel zone localization, wherein application of the wheel zone localization comprises:
determining a real-world wheel center position of the road wheel based on signals generated by a sensing-based device of the host vehicle;
mapping the real-world center position of the road wheel to a wheel center position of the road wheel in the captured image;
regenerating the captured image so that an optical axis of the captured image is perpendicular to a plane of a face of the road wheel;
identifying a diameter of a tire of the road wheel in the regenerated image; and
generating a localized wheel zone, the localized wheel zone being sized as a function of the diameter of the tire;

determining whether or not precipitation is present in the region of interest; and
generating a wet road surface signal in response to a determination that precipitation is present in the region of interest.

8. The method of claim 7 , wherein regenerating the captured image so that the optical axis of the captured image is perpendicular to the plane of the face of the road wheel comprises regenerating the captured image until the road wheel in the captured image displays a shape of a circle.

9. The method of claim 8 , wherein regenerating the captured image until the road wheel in the captured image displays the shape of a circle includes generating a synthetic image as taken from a virtual camera having an optical axis that is perpendicular to a planar face of the road wheel.

10. The method of claim 9 , wherein generating the synthetic image comprises:
identifying a plurality of pixels in the captured image;
determining a respective correlation between each of the pixels in the captured image and a respective one of a plurality of pixels in the virtual image as viewed by the optical axis perpendicular to the planar face of the road wheel; and
mapping each of the pixels in the captured image to the correlated respective one of the pixels in the virtual image.

11. The method of claim 7 , wherein the sizing the localized wheel zone as a function of the diameter of the tire includes sizing the localized wheel zone as a square, wherein each side of the square is sized to a predetermined value times the diameter of the tire, wherein the square is centered at the wheel center position of the wheel.

12. A method for determining a wet surface condition of a road with first and second lanes, the method comprising:
capturing an image of a road wheel of a remote vehicle traveling in the first lane by an image capture device of a host vehicle traveling in the second lane;
identifying, in the captured image by a processor of the host vehicle, a region of interest relative to the road wheel of the remote vehicle;
determining whether or not precipitation is present in the region of interest in the captured image, including:
analyzing the region of interest as a gray-level image;
applying a filter to the captured image to identify noise in the gray-level image, the noise representing precipitation in the region of interest; and
determining whether or not precipitation is present in the region of interest based on a non-uniformity of color in the filtered image; and

generating a wet road surface signal in response to a determination that precipitation is present in the region of interest.

13. The method of claim 12 , wherein the filter includes a median filter that generates a filtered image of the captured image that includes no noise, and wherein a noise image is generated as function of a difference between an unfiltered image of the captured image and the filtered image.

14. The method of claim 12 , further comprising applying a binary conversion to the filtered image to determine whether or not precipitation is present, the binary conversion converting precipitation to a white color.

15. The method of claim 1 , wherein the wet road surface signal includes an alert to a driver of the host vehicle of a potential hydroplaning.

16. The method of claim 1 , wherein the wet road surface signal includes an alert to a driver of the host vehicle of a potential reduced traction between one or more vehicle tires of the host vehicle and the road surface.

Assignees

• Gm Global Tech Operations Llc

Inventors

• Zhao Qingrong
• Wang Jinsong
• Zhang Qi
• Litkouhi Bakhtiar B

Classifications

• B60T8/171

  Detecting parameters used in the regulation; Measuring values used in the regulation · CPC title

• B60T2210/13

  Aquaplaning, hydroplaning · CPC title

• G01N21/84 Primary

  Systems specially adapted for particular applications · CPC title

• B60W40/064

  Degree of grip · CPC title

• B60R16/0237

  circuits concerning the atmospheric environment (meteorology G01W) · CPC title