ROAD LANE BOUNDARY DETECTION SYSTEM AND ROAD LANE BOUNDARY DETECTING METHOD

• US 20090296987A1
• Filed: 05/27/2009
• Published: 12/03/2009
• Est. Priority Date: 05/27/2008
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Inventors: Kageyama, Shioya, Tamaoki, Tomoyasu
• Granted Patent: US 8,090,152 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• 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

Abstract

A road lane boundary detection system includes a detection region setting unit that sets a certain region in a road image, as a target detection region to be searched for detection of a road lane boundary, and a detecting unit that processes image data in the target detection region set by the detection region setting unit, so as to detect the road lane boundary. The detection region setting unit sets a first detection region as the target detection region if no road lane boundary is detected, and sets a second detection region as the target detection region if the road lane boundary is detected, such that the first and second detection regions are different in size from each other.

Claims

1. A road lane boundary detection system for detecting a road lane boundary provided on a road, from a road image that is a captured image of a road ahead of a vehicle, comprising:

a detection region setting unit that sets a certain region in the road image, as a target detection region to be searched for detection of the road lane boundary; and

a detecting unit that processes image data in the target detection region set by the detection region setting unit and that detects the road lane boundary,

wherein the detection region setting unit sets a first detection region as the target detection region if no road lane boundary is detected, and sets a second detection region as the target detection region if the road lane boundary is detected, such that the first detection region and the second detection region are different in size from each other.

2. The road lane boundary detection system according to claim 1, wherein the detection region setting unit sets the second detection region such that a width of the second detection region is smaller than that of the first detection region.

3. The road lane boundary detection system according to claim 2, wherein the detection region setting unit comprises:

an initial detection region setting unit that sets, in the road image, an initial detection region of a certain size which is positioned at a first distance from the vehicle in real space and has a first dimension as measured in a width direction of the road, as the target detection region;

a close detection region setting unit that sets a close detection region that is positioned at a distance substantially equal to the first distance from the vehicle in real space, and has a second dimension as measured in the road width direction, which is smaller than the first dimension of the initial detection region, as the target detection region; and

a remote detection region setting unit that sets a remote detection region that is positioned at a second distance that is larger than the first distance from the vehicle in real space, and has a third dimension as measured in the road width direction, which is smaller than the first dimension of the initial detection region, as the target detection region, and

wherein the detecting unit processes image data in the initial detection region if no road lane boundary is detected, so as to detect a road lane boundary, and processes image data in the close detection region and the remote detection region if the road lane boundary is detected, so as to detect the road lane boundary.

4. The road lane boundary detection system according to claim 3, wherein the third dimension of the remote detection region is smaller than the second dimension of the close detection region in the road image.

5. The road lane boundary detection system according to claim 3, wherein the initial detection region is set at a position about 7 m apart from the vehicle in real space.

6. The road lane boundary detection system according to claim 3, wherein:

the close detection region setting unit sets a position of the close detection region as viewed in the road width direction in the road image, based on a position of the road lane boundary detected by the detecting unit using the initial detection region; and

the remote detection region setting unit sets a position of the remote detection region as viewed in the road width direction in the road image, based on the position of the road lane boundary detected by the detecting unit using the initial detection region.

7. The road lane boundary detection system according to claim 6, wherein:

the close detection region setting unit sets the position of the close detection region in the road image, such that the road lane boundary detected by the detecting unit using the initial detection region is located at a center of the close detection region as viewed in the road width direction; and

the remote detection region setting unit sets the position of the remote detection region in the road image, such that the road lane boundary detected by the detecting unit using the initial detection region is located at a center of the remote detection region as viewed in the road width direction.

8. The road lane boundary detection system according to claim 1, wherein the target detection region comprises a pair of regions that are separated from each other in the road width direction, and each of the regions has a width that contains each of left and right lane boundaries located to the left and right of the vehicle.

9. The road lane boundary detection system according to claim 1, wherein the road lane boundary comprises Botts dots.

10. A road lane boundary detecting method for detecting a road lane boundary provided on a road, from a road image that is a captured image of a road ahead of a vehicle, comprising:

setting a certain region in the road image, as a target detection region to be searched for detection of the road lane boundary; and

processing image data in the target detection region and detecting the road lane boundary,

wherein a first detection region is set as the target detection region for use in a condition where no road lane boundary has been detected, and a second detection region is set as the target detection region for use in a condition where the road lane boundary has been detected, such that the first detection region and the second detection region are different in size from each other.