Environment recognition device and environment recognition method

• US 8,625,850 B2
• Filed: 05/16/2012
• Issued: 01/07/2014
• Est. Priority Date: 05/19/2011
• Status: Active Grant

• Current Assignee: Subaru Corp.
• Original Assignee: Fuji Heavy Industries Limited (Subaru Corp.)
• Inventors: Kasaoki, Seisuke
• Primary Examiner(s): Carter, Aaron W
• Application Number:
  US13/472,788
• Publication Number:
  US 20120294485A1
• Time in Patent Office: 601 Days
• Field of Search: 382/103, 382/104, 382/106, 356/3, 356/11, 348118-120, 348135-142, 348/148, 348/169
• US Class Current: 382/103
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes

Abstract

There are provided an environment recognition device and an environment recognition method. The device obtains position information of a target portion in a detection area, including a relative distance from a subject vehicle; groups continuous target portions into a target object of which position differences in a width direction vertical to an advancing direction of the vehicle and in a depth direction parallel to the advancing direction fall within a first distance; determines that the target object is a candidate of a wall, when the target portions forming the target object forms a tilt surface tilting at a predetermined angle or more with respect to a plane vertical to the advancing direction; and determines that the continuous wall candidates of which position differences in the width and depth directions among the wall candidates fall within a second predetermined distance longer than the first predetermined distance are a wall.

Claims

1. An environment recognition device comprising:

a position information obtaining unit that obtains position information of a target portion in a detection area of a luminance image, including a relative distance from a subject vehicle;

a grouping unit that groups target portions, of which positions differences in a vertical direction and a parallel direction with respect to a base line which corresponds to advancing direction of the subject vehicle fall within a first distance, into a target object;

a candidate determining unit that determines that the target object is a candidate of a wall, when the target portions included in the target object form a tilt surface tilting at a predetermined angle or more against a vertical plane with respect to a base plane which corresponds to the advancing direction of the subject vehicle; and

a wall determining unit that determines that the wall candidates, of which positions differences in the vertical direction and the parallel direction with respect to a base line which corresponds to the advancing direction of the subject vehicle fall within a second predetermined distance longer than the first predetermined distance, are a wall.

2. The environment recognition device according to claim 1, wherein the grouping unit makes a determination for the grouping based on not only the position differences in the vertical direction and the parallel direction with respect to a base line which corresponds to the advancing direction of the subject vehicle but also a position difference in the height direction from a road surface, and the wall determining unit may make a determination of a wall based on not only the position differences in the vertical direction and in the parallel direction with respect to a base line which corresponds to the advancing direction of the vehicle but also the position difference in the height direction from the road surface.

3. The environment recognition device according to claim 1, further comprising:

an environment determining unit that determines whether or not the detection area is in an environment where light is diffused, wherein

the wall determining unit determines all of the wall candidates as a wall when the environment determining unit determines that the detection area is in the environment where light is not diffused.

4. The environment recognition device according to claim 1, further comprising:

a distance deriving unit that derives each average of in-line relative distances, which are assigned to target portions included in a target object and which are aligned in a horizontal direction of the image corresponding to the width direction in the real world; and

a group dividing unit that divides the target object at a position where differences of the averages in the vertical direction of the image corresponding to the height direction in the real world exceeds the predetermined threshold value, and defines the position comprised of the horizontally aligned target portions as a boundary of the divided target objects.

5. The environment recognition device according to claim 4, wherein the group dividing unit determines the predetermined threshold value based on the average of the relative distances of all target portions included in the target object.

7. The environment recognition device according to claim 2, further comprising:

an environment determining unit that determines whether or not the detection area is in an environment where light is diffused, wherein

the wall determining unit determines all of the wall candidates as a wall when the environment determining unit determines that the detection area is in the environment where light is not diffused.

8. The environment recognition device according to claim 2, further comprising:

a distance deriving unit that derives each average of in-line relative distances, which are assigned to target portions included in a target object and which are aligned in a horizontal direction of the image corresponding to the width direction in the real world; and

a group dividing unit that divides the target object at a position where differences of the averages in the vertical direction of the image corresponding to the height direction in the real world exceeds the predetermined threshold value, and defines the position comprised of the horizontally aligned target portions as a boundary of the divided target objects.

9. The environment recognition device according to claim 3, further comprising:

a distance deriving unit that derives each average of in-line relative distances, which are assigned to target portions included in a target object and which are aligned in a horizontal direction of the image corresponding to the width direction in the real world; and

a group dividing unit that divides the target object at a position where differences of the averages in the vertical direction of the image corresponding to the height direction in the real world exceeds the predetermined threshold value, and defines the position comprised of the horizontally aligned target portions as a boundary of the divided target objects.

10. The environment recognition device according to claim 8, wherein the group dividing unit determines the predetermined threshold value based on the average of the relative distances of all target portions included in the target object.

11. The environment recognition device according to claim 9, wherein the group dividing unit determines the predetermined threshold value based on the average of the relative distances of all target portions included in the target object.

6. An environment recognition method comprising:

obtaining position information of a target portion in a detection area of a luminance image including a relative distance from a subject vehicle;

grouping target portions, of which positions differences in a vertical direction and a parallel direction with respect to a base line which corresponds to advancing direction of the subject vehicle fall within a first predetermined distance, into a target object;

determining that the target object is a candidate of a wall, when the target portions included in the target object form a tilt surface tilting at a predetermined angle or more against a vertical plane with respect to a base plane which corresponds to the advancing direction of the subject vehicle; and

determining that determines that the wall candidates, of which positions differences in the vertical direction and the parallel direction with respect to a base line which corresponds to the advancing direction of the subject vehicle fall within a second predetermined distance longer than the first predetermined distance, are a wall.