Method and device that recognizes road users in an environment of a vehicle

Patent metadata

Field	Value
Publication number	US-10831194-B2
Application number	US-201816037540-A
Country	US
Kind code	B2
Filing date	Jul 17, 2018
Priority date	Jul 25, 2017
Publication date	Nov 10, 2020
Grant date	Nov 10, 2020

Abstract

The disclosure relates to a method and to a device that recognizes road users in an environment of a vehicle. The vehicle includes at least one camera-based sensor that acquires images of the vehicle environment, wherein the device implements the method to process the images. The method includes acquiring at each of a plurality of different times, an image of an object located in the vehicle environment, performing image analyses on the images acquired at different times, and deciding, based on a result of the image analyses performed, whether the object shall be interpreted as a road user or as an inanimate entity. The device implements the method to control an autonomous driving system to control vehicle movements based on the result.

First claim

What is claimed is:

1. A method that recognizes road users in an environment of a vehicle comprising:
acquiring images, via a camera, of the vehicle environment and, at each of a plurality of different times, of an object located in the vehicle environment;
controlling a vehicle movement according whether an object within the images is interpreted as a road user or as an inanimate entity according to image analyses performed on the images acquired at the different times, wherein performing image analyses includes checking whether the images acquired at different times exhibit at least one feature indicating an existence of an advertisement;
performing the checking the images acquired at different times by accessing at least one database stored in a Cloud memory; and
running at least one character recognition algorithm or object recognition algorithm during performing the checking of the images acquired at the different times.

2. The method as claimed in claim 1 , wherein performing image analyses includes applying a similarity transformation to any two images of an object that are acquired at consecutive of the different times.

3. The method as claimed in claim 1 further comprising analyzing an infrared signature of the object, wherein the result is based on the infrared signature of the object.

4. The method as claimed in claim 1 further comprising analyzing a light spectrum of the images acquired at different times, wherein whether the object within the images is interpreted as a road user or as an inanimate entity is further based on the light spectrum.

5. The method as claimed in claim 1 , wherein controlling a vehicle movement includes transmitting a result of whether the object within the images is interpreted as a road user or as an inanimate entity to an autonomous-driving system that controls vehicle components.

6. A vehicle autonomous driving system comprising:
at least one camera-based sensor that acquires images of a vehicle environment; and
a device configured to, in response to a result indicative of a road user or inanimate entity from an image analyses of an image acquired at each of a plurality of different times, via the at least one camera-based sensor, of an object located in the vehicle environment, control a vehicle movement according to the result, wherein the image analyses include a check of the image acquired at each of the plurality of different times of the object for at least one feature indicative of an advertisement, and wherein the check is performed via access to at least one database stored in a Cloud memory and the check runs at least one character recognition algorithm or object recognition algorithm.

7. The autonomous driving system as claimed in claim 6 , wherein the result is based on an infrared signature of the object taken by the at least one camera-based sensor.

8. The autonomous driving system as claimed in claim 6 , wherein the result is based a light spectrum of the images acquired at a plurality of different times taken by the at least one camera-based sensor.

9. A vehicle comprising:
an autonomous driving system having a camera that acquires images of an environment; and
a controller configured to, in response to a result indicative of a road user or inanimate entity from an image analyses of an image acquired at each of a plurality of different times, via the camera, of an object located in the vehicle environment, control, via the autonomous driving system, a vehicle movement according to the result, wherein the image analyses includes a check of the image acquired at each of the plurality of different times of the object for at least one feature indicative of an advertisement, and wherein the check is performed via access to at least one database stored in a Cloud memory and the check runs at least one character recognition algorithm or object recognition algorithm.

10. The vehicle as claimed in claim 9 , wherein the image analyses include application of a similarity transformation to any two images of an object that are acquired at consecutive times.

11. The vehicle as claimed in claim 9 , wherein the controller is configured to transmit the result to the autonomous driving system that controls vehicle components according to the result.

12. The vehicle as claimed in claim 9 , wherein the result is based on an infrared signature of the object taken by the at least one camera-based sensor.

13. The vehicle as claimed in claim 9 , wherein the result is based a light spectrum of the images acquired at a plurality of different times taken by the at least one camera-based sensor.

Assignees

• Ford Global Tech Llc

Inventors

• Stefan Frederic
• Arndt Christoph
• Gussen Uwe

Classifications

• G06V30/10

  Character recognition · CPC title

• G06V20/58 Primary

  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

• G06V20/56

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

• G06V20/62

  Text, e.g. of license plates, overlay texts or captions on TV images · CPC title

• G06F16/5854

  using shape and object relationship · CPC title