Systems and methods for determining the status and details of a traffic light

Patent metadata

Field	Value
Publication number	US-2016306361-A1
Application number	US-201615197112-A
Country	US
Kind code	A1
Filing date	Jun 29, 2016
Priority date	Jan 30, 2014
Publication date	Oct 20, 2016
Grant date	—

Abstract

Systems and methods use cameras to provide autonomous navigation features. In one implementation, a traffic light detection system is provided for a vehicle. One or more processing devices associated with the system receive at least one image of an area forward of the vehicle via a data interface, with the area including at least one traffic lamp fixture having at least one traffic light. The processing device(s) determine, based on at least one indicator of vehicle position, whether the vehicle is in a turn lane. Also, the processing device(s) process the received image(s) to determine the status of the traffic light, including whether the traffic light includes an arrow. Further, the system may cause a system response based on the determination of the status of the traffic light, whether the traffic light includes an arrow, and whether the vehicle is in a turn lane.

First claim

1 - 25 . (canceled)

26 . A traffic light detection system for a vehicle, the system comprising:
a data interface configured to receive a plurality of images from an image capture device configured to acquire images of an environment forward of the vehicle, the environment including a traffic lamp fixture having at least one traffic light; and at least one processing device configured to:
receive the plurality of images via the data interface;
align at least two of the plurality of images based on a center of brightness;
increase a number of pixels associated with a portion of at least one of the aligned images, wherein the portion corresponds to an area of the at least one of the aligned images in which image data corresponding to the traffic light appears;
determine pixel values associated with the increased number of pixels; and
determine, based on the determined pixel values, at least one characteristic of the traffic light.

27 . The traffic light detection system of claim 26 , wherein the at least one characteristic is at least one of a shape, color, and position.

28 . The traffic light detection system of claim 26 , wherein increasing the number of pixels associated with the portion of at least one of the aligned images includes dividing at least one pixel associated with the portion into a plurality of pixels.

29 . The traffic light detection system of claim 26 , wherein aligning the at least two of the plurality of images includes rotating at least one of the at least two of the plurality of images.

30 . The traffic light detection system of claim 26 , wherein increasing the number of pixels associated with the portion of at least one of the aligned images includes increasing the number of pixels associated with the portion of at least one of the aligned images such that the aligned images are scaled to a predetermined size.

31 . The traffic light detection system of claim 30 , wherein increasing the number of pixels associated with the portion of at least one of the aligned images includes determining a scaling factor based on a size of an object in the at least one of the aligned images and the predetermined size.

32 . The traffic light detection system of claim 26 ,
wherein the at least one characteristic of the traffic light includes a shape of a visual component of the traffic light, and wherein the at least one processing device is further configured to determine whether the shape is an arrow.

33 . The traffic light detection system of claim 32 , wherein the at least one processing device is further configured to determine a status of the traffic light.

34 . The traffic light detection system of claim 33 , wherein determining a status of the traffic light includes any of determining whether the traffic light is illuminated, determining a position of the traffic light within the traffic lamp fixture, and determining a color associated with the traffic light.

35 . The traffic light detection system of claim 33 , wherein the at least one processing device is further configured to determine, based on the determination of whether the shape is an arrow and at least one indicator of vehicle position, whether the vehicle is in a turn lane.

36 . The traffic light detection system of claim 35 , wherein the at least one processing device is further configured to cause a system response based on the determination of the status of the traffic light, the determination of whether the shape is an arrow, and whether the vehicle is in a turn lane.

37 . The traffic light detection system of claim 32 , wherein the at least one indicator of vehicle position includes a lane marker recognized based on analysis of at least one of the plurality of images.

38 . The traffic light detection system of claim 36 , wherein the system response includes any of providing a notice to the driver of the vehicle, applying vehicle brakes, discontinuing cruise control, and initiating an automated turning maneuver.

39 . The traffic light detection system of claim 35 , wherein the at least one indicator of vehicle position includes a distance to an Intersection.

40 . The traffic light detection system of claim 35 , wherein increasing the number of pixels includes expanding the number of pixels by replacing a pixel with a matrix of pixels.

41 . A vehicle, comprising:
a body; a data interface configured to receive a plurality of images from an image capture device configured to acquire images of an environment forward of the vehicle, the environment including a traffic lamp fixture having at least one traffic light; and at least one processing device configured to:
receive the plurality of images via the data interface;
align at least two of the plurality of images based on a center of brightness;
increase a number of pixels associated with a portion of at least one of the aligned images, wherein the portion corresponds to an area of the at least one of the aligned images in which image data corresponding to the traffic light appears;
determine pixel values associated with the increased number of pixels; and
determine, based on the determined pixel values, at least one characteristic of the traffic light.

42 . The vehicle of claim 41 , wherein the at least one characteristic is at least one of a shape, color, and position.

43 . The vehicle of claim 41 , wherein increasing the number of pixels associated with the portion of at least one of the aligned images includes dividing at least one pixel associated with the portion into a plurality of pixels.

44 . The vehicle of claim 41 , wherein aligning the at least two of the plurality of images includes rotating at least one of the at least two of the plurality of images.

45 . The vehicle of claim 41 ,
wherein the at least one characteristic of the traffic light includes a shape of a visual component of the traffic light, and wherein the at least one processing device is further configured to determine whether the shape is an arrow.

46 . The vehicle of claim 41 , wherein increasing the number of pixels includes expanding the number of pixels by replacing a pixel with a matrix of pixels.

47 . A method for traffic light detection, comprising:
receiving, from an image capture device, a plurality of images; aligning at least two of the plurality of images based on a center of brightness; increasing a number of pixels associated with a portion of at least one of the aligned images, wherein the portion corresponds to an area of the at least one of the aligned images in which image data corresponding to a traffic light appears; determining pixel values associated with the increased number of pixels; and determining, based on the determined pixel values, at least one characteristic of the traffic light.

Assignees

• Mobileye Vision Technologies Ltd

Inventors

• Ben Shalom Itai
• Shalev-Shwartz Shai
• Bagon Eyal
• Taieb Yoav
• Shashua Amnon
• Huberman David

Classifications

• G08G1/167

  Driving aids for lane monitoring, lane changing, e.g. blind spot detection · CPC title

• G08G1/165

  for passive traffic, e.g. including static obstacles, trees · CPC title

• B60W2554/4041

  Position · CPC title

• B60W2555/60

  Traffic rules, e.g. speed limits or right of way · CPC title

• B60W2552/30

  Road curve radius · CPC title