Vehicle-mounted environment recognition apparatus and vehicle-mounted environment recognition system

• US 9,064,418 B2
• Filed: 08/13/2010
• Issued: 06/23/2015
• Est. Priority Date: 03/17/2010
• Status: Active Grant

• Current Assignee: Hitachi Astemo Ltd. (Hitachi, Ltd.)
• Original Assignee: Hitachi Automotive Systems Limited (Hitachi, Ltd.)
• Inventors: Takemura, Masayuki, Muramatsu, Shoji, Furusawa, Isao, Shima, Takeshi, Otsuka, Yuji, Monji, Tatsuhiko, Ohtsuji, Shinya
• Primary Examiner(s): Natnael, Paulos M
• Application Number:
  US13/575,173
• Publication Number:
  US 20120327188A1
• Time in Patent Office: 1,775 Days
• Field of Search: 348/61, 348/149, 348/112, 348/120, 348/148, 348/169, 348/116, 348/118, 348/119, 348/113, 382/103, 382/104, 382/106, 382/107
• US Class Current: 1/1
• CPC Class Codes: G01S 11/12 using electromagnetic waves other than radio waves
• CPC Class Codes: G06V 20/584 of vehicle lights or traffic lights
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
• CPC Class Codes: G08G 1/16 Anti-collision systems road vehicle drive control systems for predicting or avoiding probable or impending collision otherwise than by control of a particular sub-unit B60W30/08
• CPC Class Codes: G08G 1/165 for passive traffic, e.g. including static obstacles, trees
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes
• CPC Class Codes: G08G 1/167 Driving aids for lane monitoring, lane changing, e.g. blind spot detection

Abstract

A vehicle-mounted environment recognition apparatus including a simple pattern matching unit which extracts an object candidate from an image acquired from a vehicle-mounted image capturing apparatus by using a pattern shape stored in advance and outputs a position of the object candidate, an area change amount prediction unit which calculates a change amount prediction of the extracted object candidate on the basis of an object change amount prediction calculation method set differently for each area of a plurality of areas obtained by dividing the acquired image, detected vehicle behavior information, and an inputted position of the object candidate, and outputs a predicted position of an object, and a tracking unit which tracks the object on the basis of an inputted predicted position of the object.

Claims

1. A vehicle-mounted environment recognition apparatus comprising:

a pattern matching unit which extracts an object candidate from an image acquired from a vehicle-mounted image capturing apparatus by using a pattern shape stored in advance and outputs a position of the object candidate;

an area change amount prediction unit which calculates a change amount prediction of the extracted object candidate on the basis of an object change amount prediction calculation method set differently for each area of a plurality of areas obtained by dividing the acquired image, detected vehicle behavior information, and an inputted position of the object candidate, and outputs a predicted position of an object;

a tracking unit which tracks the object on the basis of an inputted predicted position of the object; and

an area setting unit which divides the acquired image into a first area including a vanishing point and an area different from the first area to set a plurality of areas.

2. The vehicle-mounted environment recognition apparatus according to claim 1, wherein

the plurality of areas set by the area setting unit are a middle layer left area and a middle layer right area obtained by dividing the first area into two areas at the vanishing point, a low layer area which is set on one side of the first area in the vertical direction in the second area, and a high layer area which is set on the other side of the first area in the vertical direction in the second area.

3. The vehicle-mounted environment recognition apparatus according to claim 1, wherein

the vanishing point is calculated based on a result of a lane recognition process performed using the acquired image.

4. The vehicle-mounted environment recognition apparatus according to claim 3, wherein

a setting of the first area set by the area setting unit is dynamically changed based on the vanishing point calculated based on the result of the lane recognition process.

5. The vehicle-mounted environment recognition apparatus according to claim 1, wherein

a setting of the first area set by the area setting unit is dynamically changed based on acquired map information.

6. The vehicle-mounted environment recognition apparatus according to claim 1, wherein

a predetermined object change amount prediction calculation method in the first area is a method which assumes a wall perpendicular to a road on which the vehicle runs and performs a prediction calculation.

7. A vehicle-mounted environment recognition apparatus comprising:

a pattern matching unit which extracts an object candidate from an image acquired from a vehicle-mounted image capturing apparatus by using a pattern shape stored in advance and outputs a position of the object candidate;

an area change amount prediction unit which calculates a change amount prediction of the extracted object candidate on the basis of an object change amount prediction calculation method set differently for each area of a plurality of areas obtained by dividing the acquired image, detected vehicle behavior information, and an inputted position of the object candidate, and outputs a predicted position of an object; and

a tracking unit which tracks the object on the basis of an inputted predicted position of the object;

wherein

the area change amount prediction unit includes

an area three-dimensional position presumption unit which presumes a three-dimensional position of the object candidate in a three-dimensional vehicle coordinate system using a position of the vehicle as an origin from the position of the object candidate in a two-dimensional image coordinate system inputted from the pattern matching unit,

an area three-dimensional position prediction unit which predicts a three-dimensional position of the object candidate in the vehicle coordinate system after predetermined periods on the basis of the vehicle behavior information and a presumption result of the three-dimensional position of the object candidate in the vehicle coordinate system obtained by the area three-dimensional position presumption unit, and

an area image position prediction unit which converts a prediction result of the three-dimensional position predicted by the area three-dimensional position prediction unit into image coordinates and calculates an image position of the object.

8. A vehicle-mounted environment recognition apparatus comprising:

a pattern matching unit which extracts an object candidate from an image acquired from a vehicle-mounted image capturing apparatus by using a pattern shape stored in advance and outputs a position of the object candidate;

an area change amount prediction unit which calculates a change amount prediction of the extracted object candidate on the basis of an object change amount prediction calculation method set differently for each area of a plurality of areas obtained by dividing the acquired image, detected vehicle behavior information, and an inputted position of the object candidate, and outputs a predicted position of an object;

a tracking unit which tracks the object on the basis of an inputted predicted position of the object; and

an object determination unit which determines whether or not the object is an on-street motionless object on the basis of a result where the tracking unit tracks the object;

wherein

if the object is determined not to be an on-street motionless object, the object determination unit determines whether or not the object is a tail light of a vehicle on the basis of the result of the tracking.