Lane marker recognition method

• US 20030103650A1
• Filed: 11/27/2002
• Published: 06/05/2003
• Est. Priority Date: 11/30/2001
• Status: Active Grant

• Current Assignee: Hitachi Astemo Ltd. (Hitachi, Ltd.)
• Original Assignee: Hitachi, Ltd.
• Inventors: Otsuka, Yuji, Muramatsu, Shoji, Takenaga, Hiroshi, Monji, Tatsuhiko, Takezaki, Jiro
• Granted Patent: US 7,362,883 B2
• Time in Patent Office: Days
• US Class Current: 382/104
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30256 Lane; Road marking
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/73 using feature-based methods
• 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

In a lane marker recognition method comprising the steps of inputting an image including a lane marker, extracting a luminance change point in the image input in the image inputting step, and estimating the position of the lane marker using the edge points extracted in the edge extracting step, the edge extracting step calculates an angle of orientation of each of the edges and the lane marker position estimating step extracts edge points that are oriented toward a vanishing point of a road from among the edge points extracted in the edge extracting step.

Claims

1. A lane marker recognition method, comprising the steps of:

inputting an image including a lane marker;

extracting an edge point, which is a luminance change point in the image input in the image inputting step, and calculating an angle representing an orientation of each of the edge points extracted; and

estimating a position of the lane marker by extracting, from among the edge points extracted in the edge extracting step, those edge points the angles of which are a predetermined value or fall within a predetermined range of values.

2. The lane marker recognition method according to claim 1, wherein the predetermined value which the angles of the edge points extracted in the lane marker position estimating step take or the predetermined range of values which the angles of the edge points extracted in the lane marker position estimating step fall within varies according to the position of each edge point.

3. The lane marker recognition method according to claim 1, wherein the predetermined value which the angles of the edge points extracted in the lane marker position estimating step take is the angle of a straight line connecting the edge point and a vanishing point of a road or the predetermined range of values which the angles of the edge points extracted in the lane marker position estimating step fall within is the range deviated 180 degrees from the angle of the straight line.

4. The lane marker recognition method according to claim 3, wherein the lane marker position estimating step eliminates edge points that fall within a predetermined range from the vanishing point.

5. The lane marker recognition method according to claim 3, wherein the lane marker position estimating step creates a histogram of edge points by angle from edge points having an angle oriented toward the vanishing point and estimates a lane marker based on the edge points having angles with a frequency of occurrence of a predetermined value or more.

6. The lane marker recognition method according to claim 4, wherein the lane marker position estimating step creates a histogram of edge points by angle from edge points having an angle oriented toward the vanishing point, of all the edge points falling outside a predetermined range from the vanishing point, and estimates a lane marker based on the edge points having angles with a frequency of occurrence of a predetermined value or more.

7. The lane marker recognition method according to claim 5, wherein the lane marker position estimating step limits all the angles with high frequencies as obtained from the histogram to the angles of left and right lane markers based on a lane width.

8. The lane marker recognition method according to claim 7, wherein the lane marker position estimating process makes linear approximation of the lane marker position from a distribution of edge points having the angles of the lane markers.

9. A lane marker recognition method, comprising the steps of:

inputting an image including a lane marker;

changing in advance a ratio of width to height of the image input in the image inputting step;

extracting an edge point, which is a luminance change point in the image changed in advance, and calculating an angle representing an orientation of each of the edge points extracted; and

estimating a position of the lane marker by extracting, from among the edge points extracted in the edge extracting step, those edge points the angles of which are a predetermined value or fall within a predetermined range of values.