Systems and methods for determining a condition of a road surface

Patent metadata

Field	Value
Publication number	US-9598087-B2
Application number	US-201414568656-A
Country	US
Kind code	B2
Filing date	Dec 12, 2014
Priority date	Dec 12, 2014
Publication date	Mar 21, 2017
Grant date	Mar 21, 2017

Abstract

Systems and methods use an image of a road surface that is generated by a camera. The image includes a pattern from a light source. A region of interest in the image is determined based on the pattern from the light source. A total area is determined that includes at least part of the region of interest and an area adjacent the region of interest. A feature vector is extracted based on the region of interest and the total area. A road condition is determined based on the feature vector and a classifier.

First claim

What is claimed is:

1. A method, comprising:
accessing, by a processor, an image of a road surface that is generated by a camera, wherein the image includes a pattern from a light source;
determining, by the processor, a region of interest in the image, the region of interest including pixels from the image, wherein the region of interest is based on the pattern from the light source;
determining, by the processor, a total area in the image, the total area including pixels from the image, wherein the total area includes at least part of the region of interest and an area adjacent the region of interest;
extracting a feature vector based on characteristics of the region of interest and the total area; and
determining a road condition by comparing the feature vector to a classifier.

2. The method of claim 1 , wherein the pixels of the region of interest are a subset of the pixels of the total area.

3. The method of claim 1 , wherein the classifier includes a boundary that represents a road condition.

4. The method of claim 3 , wherein the classifier includes a boundary that represents a road condition that is one of an ice-covered surface and a water-covered surface.

5. The method of claim 3 , wherein the classifier is generated using one of a support vector machine technique and a linear discriminant analysis technique.

6. The method of claim 1 , further comprising converting, by the processor, the region of interest and the total area to one of: binary values; and values generated using a Laplacian of Gaussian filter.

7. The method of claim 1 , wherein the extracting a feature vector includes:
extracting a first set of characteristics from the region of interest;
extracting a second set of characteristics from the total area; and
generating a feature vector based on the first set of characteristics and the second set of characteristics.

8. The method of claim 7 , wherein:
the first set of characteristics includes a mean of intensity of the region of interest;
the second set of characteristics includes a mean of intensity of the total area; and
the feature vector includes a first feature that is the mean of intensity of the region of interest divided by the mean of intensity of the total area.

9. The method of claim 7 , wherein:
the first set of characteristics includes a variance of intensity of the region of interest; and
the feature vector includes a first feature that is the variance of intensity of the region of interest.

10. The method of claim 7 , wherein:
the first set of characteristics includes a variance of intensity of the region of interest;
the second set of characteristics includes a variance of intensity of the total area; and
the feature vector includes a first feature that is the variance of intensity of the region of interest divided by the variance of intensity of the total area.

11. The method of claim 7 , the feature vector includes a first feature that is a maximum response of the region of interest after being processed by a Laplacian of Gaussian filter.

12. The method of claim 1 , wherein extracting a feature vector includes selecting a subset of features using a principle component analysis technique.

13. The method of claim 1 , wherein the road condition is a first road condition, the method further comprising:
accessing a second road condition; and
determining an overarching road condition indication based on at least the first road condition and the second road condition.

14. The method of claim 13 , wherein the region of interest is a first region of interest;
wherein the first road condition is based on the first region of interest;
wherein the second road condition is based on a second region of interest;
and wherein the first region of interest and the second region of interest are distributed over at least one of time and space.

15. The method of claim 14 , wherein the first region of interest and the second region of interest include different pixels in the image.

16. The method of claim 15 , wherein the first region of interest and the second region of interest are selected based on the pattern of the light source.

17. The method of claim 14 , wherein the first region of interest and the second region of interest are in different images, wherein the different images are at least one of generated at different times and generated by different cameras.

18. A system, for use with a vehicle, comprising:
a light source configured to provide a light pattern on a road surface;
a camera configured to generate an image of the road surface, the image including the light pattern;
a processor; and
a memory, comprising:
instructions that, when executed by the processor, cause the processor to perform operations, the operations comprising:
accessing an image of a road surface that is generated by the camera, the image including pixels, each of the pixels having an intensity value;
determining a region of interest from the image, the region of interest including pixels from the image;
determining a total area from the image, the total area including pixels from the image, wherein the total area includes pixels in the region of interest and pixels in an area of the image that is adjacent of the region of interest;
extracting a feature vector based on the region of interest and the total area; and

determining a road condition based on the feature vector and a classifier.

19. The system of claim 18 , the extracting a feature vector based on the region of interest and the total area including:
extracting a first set of characteristics from the region of interest; and
extracting a second set of characteristics from the total area.

20. A non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to perform operations, the operations comprising:
accessing an image of a road surface that is generated by a camera, the image including pixels, each of the pixels having an intensity value;
determining a region of interest from the image, the region of interest including pixels from the image;
determining a total area from the image, the total area including pixels from the image, wherein the total area includes pixels in the region of interest and pixels in an area of the image adjacent the region of interest;
extracting a feature vector based on the region of interest and the total area; and
determining a road condition based on the feature vector and a classifier.

Assignees

• Gm Global Tech Operations Llc

Inventors

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

Classifications

• H04N23/56

  provided with illuminating means · CPC title

• G06F18/24

  Classification techniques · CPC title

• B60W40/06 Primary

  Road conditions · CPC title

• H04N5/2256

  Electricity · mapped topic

• G06K9/00791

  Physics · mapped topic