Method and device for detecting road region as well as method and device for detecting road line

• US 9,378,424 B2
• Filed: 09/11/2014
• Issued: 06/28/2016
• Est. Priority Date: 12/22/2011
• Status: Active Grant

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: You, Ganmei, Zheng, Jichuan
• Primary Examiner(s): Seth, Manav
• Application Number:
  US14/483,771
• Publication Number:
  US 20140379257A1
• Time in Patent Office: 656 Days
• US Class Current: 1/1
• CPC Class Codes: G01C 11/04 Interpretation of pictures
• CPC Class Codes: G01C 21/28 with correlation of data from several navigational instruments
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 2207/30256 Lane; Road marking
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06V 10/443 by matching or filtering
• 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: G06V 20/647 by matching two-dimensional images to three-dimensional objects

Abstract

Disclosed are a road line detection method and a road line detection device. The road line detection method comprises a step of obtaining a first disparity map including one or more road regions and a corresponding V-disparity image; a step of sequentially detecting plural sloped line segments in the corresponding V-disparity image according to a big-to-small order of disparities and a big-to-small order of V-values, to serve as plural sequentially adjacent road surfaces; a step of obtaining a second disparity map of plural road line regions of interest corresponding to the plural sloped line segments; and a step of detecting one or more road lines in the second disparity map of the plural road line regions of interest.

Claims

1. A road region detection method comprising:

obtaining a set of first values and a set of second values representing distances related to one or more road regions from a map including the one or more road regions and an image corresponding to the map, respectively; and

sequentially detecting plural sequentially adjacent road surfaces by iteratively detecting road surfaces based on corresponding sloped segments from a starting point where the first value and the second value are maximum, each road surface being formed by linear fitting plural reliable points, the plural reliable points being points that do not form a vertical line and, whose gray levels are greater than a threshold and first values are less than reliable points of previously detected road surfaces.

2. The method according to claim 1, further comprising:

for each of the detected road surfaces, obtaining a set of third values representing distances; and

detecting one or more road lines based on the set of third values.

3. The method according to claim 2, wherein:

the set of first values are disparities of a first disparity map including the one and more road regions;

the set of second values are V-values of a V-disparity image corresponding to the first disparity map, the V-disparity image being a lateral view of the first disparity map; and

the sequentially detecting plural sequentially adjacent road surfaces is carried out with respect to the V-disparity image.

4. The method according to claim 2, wherein:

the set of third values refers to disparities of a second disparity map.

5. The method according to claim 2, wherein:

the gray level of each of the plural reliable points forming the corresponding sloped line segment associated with a road line detected based on the set of third values is set to a same level.

6. The method according to claim 3, further comprising:

for each of the road lines detected based on the set of third values, obtaining points in a U-disparity image corresponding to a respective one of the road lines, the U-disparity image being a top view of the first disparity image;

determining whether the obtained points are located on a non-vertical and non-horizontal sloped line; and

discarding the respective one of the road lines if it is determined that the obtained points are not located on the sloped line.

7. The method according claim 3, wherein, the detecting a first road surface includes:

finding all of the sloped line segments in the corresponding V-disparity image by utilizing a Hough transform; and

selecting, from all of the sloped lines segments, one sloped line segment on which the number of pass points is maximum or whose disparity is maximum or whose length is maximum, and then setting the selected sloped line segment as the first sloped line.

8. The method according to claim 3, wherein:

the first disparity map is obtained by performing calculation with regard to left and right images of a scene in front of a vehicle; and

the detecting a first road surface includes a step of seeking points satisfying

h/bΔ

=f sin θ

+V cos θ

so as to obtain the first sloped line, wherein, h refers to a height from a twin-lens camera for capturing the left and right images to a road surface, b refers to a distance between the centers of two lenses in the twin-lens camera, θ

refers to an angle between an image surface and the road surface, f refers to a focal length of the twin-lens camera, Δ

refers to a disparity of a point on a sloped line segment, and V refers to a V-value of the point on the sloped line segment.

9. The method according to claim 3, further comprising:

before detecting the plural sloped line segments, removing at least one vertical line, whose length is greater than a predetermined threshold value, in the corresponding V-disparity image.

10. A road region detection device comprising:

a processor configured to execute program instructions, the program instructions configuring the processor to,

obtain a set of first values and a set of second values representing distances related to one or more road regions from a map including the one or more road regions and an image corresponding to the map, respectively; and

sequentially detect plural sequentially adjacent road surfaces by iteratively detecting road surfaces based on corresponding sloped segments from a starting point where the first value and the second value are maximum, each road surface being formed by linear fitting plural reliable points that do not form a vertical line as well as whose gray levels are greater than a threshold and first values are less than reliable points of previously detected road surfaces.