Method and arrangement for assessing the roadway surface being driven on by a vehicle

Patent metadata

Field	Value
Publication number	US-10467482-B2
Application number	US-201715435860-A
Country	US
Kind code	B2
Filing date	Feb 17, 2017
Priority date	Feb 17, 2016
Publication date	Nov 5, 2019
Grant date	Nov 5, 2019

Abstract

The disclosure relates to a method and an arrangement for assessing the roadway surface being driven on by a vehicle. In a method according to the disclosure, on the basis of at least one image recorded with a camera that is present on the vehicle, the roadway surface being driven on by the vehicle is classified using a classifier. The classifier is trained on the basis of image features that are extracted from the at least one image, wherein a plurality of image details are defined in the at least one image. The extraction of image features is performed independently for each of these image details.

First claim

What is claimed is:

1. A method for assessing a roadway surface for a vehicle comprising:
capturing an image of a roadway with a vehicle-mounted camera, the image defining a plurality of image details;
extracting image features independently for each of the plurality of image details, wherein at least some of the features are textural details based on an entropy of the image;
classifying the roadway surface using a classifier to determine a roadway class, the classifier including the image features as input data, wherein the classifier is trained based on the extracted image features;
determining weather conditions with at least one on-board vehicle sensor; and
utilizing a driver-assistance system to control the vehicle according to a friction map that is based the roadway class and the weather conditions.

2. The method as claimed in claim 1 further comprising weighing the image features extracted for each of the plurality of image details using different weights.

3. The method as claimed in claim 1 , wherein the classifying of the roadway surface takes place in real time.

4. The method as claimed in claim 1 further comprising checking the classification of the roadway surface based on validation data.

5. The vehicle as claimed in claim 1 , wherein the friction map indicates a coefficient of tire friction between the roadway and tires of the vehicle.

6. A vehicle arrangement for assessing the roadway surface comprising:
a camera disposed on the vehicle and configured to capture at least one image of a roadway, the image defining a plurality of image details; and
a processing unit configured to:
extract image features independently for each of the plurality of image details, wherein at least some of the features are textural details based on an entropy of the image,
classify the roadway surface using a classifier to determine a roadway class, the classifier including the image features as input data,
determining weather conditions proximate the vehicle, and
control a driver-assistance system of the vehicle according to a friction map that is based the roadway class and the weather conditions.

7. The vehicle arrangement as claimed in claim 6 , wherein the processing unit is further configured to weight each of the image features differently via the classifier.

8. The vehicle arrangement as claimed in claim 6 , wherein the processing unit classifies the roadway surface in real time.

9. The vehicle as claimed in claim 6 , wherein the camera is a black-and-white camera.

10. The vehicle as claimed in claim 6 , wherein the friction map indicates a coefficient of tire friction between the roadway and tires of the vehicle.

11. The vehicle as claimed in claim 6 , wherein at least some of the textural details are based on a gray-value matrix of the image.

12. The vehicle as claimed in claim 6 , wherein the classifier is trained based on the extracted image features.

13. A vehicle comprising:
a camera that records an image having image details of a roadway; and
a processor configured to,
extract image features for each of the image details, wherein at least some of the features are textural details based on a gray-value matrix of the image,
classify the roadway surface using a classifier to determine a roadway class, the classifier including the image features as input data,
determining weather conditions proximate the vehicle, and
control a driver-assistance system of the vehicle according to a friction map that is based the roadway class and the weather conditions.

14. The vehicle as claimed in claim 13 , wherein at least some of the textural details are based on an entropy of the image.

15. The vehicle as claimed in claim 13 , the processor classifies the roadway in real time.

16. The vehicle as claimed in claim 13 , wherein the friction map indicates a coefficient of tire friction between the roadway and tires of the vehicle.

17. The vehicle as claimed in claim 13 , wherein the camera is a black-and-white camera.

18. The vehicle as claim in claim 13 , wherein the classifier is trained based on the extracted image features.

Assignees

• Ford Global Tech Llc

Inventors

• Lakehal-Ayat Mohsen
• Potma Joost
• Seemann Michael

Classifications

• G06V10/82 Primary

  using neural networks · CPC title

• G06V10/454

  Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN] · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06V20/56

  exterior to a vehicle by using sensors mounted on the vehicle · CPC title

• G06F18/214 Primary

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title