Method and device for detecting the course of a traffic lane

• US 8,442,273 B2
• Filed: 10/18/2008
• Issued: 05/14/2013
• Est. Priority Date: 10/31/2007
• Status: Active Grant

• Current Assignee: ADC GmbH (TE Connectivity Ltd. (United Kingdom))
• Original Assignee: ADC GmbH (TE Connectivity Ltd. (United Kingdom))
• Inventors: Walter, Michael, Klein, Wladimir
• Primary Examiner(s): Tabatabai, Abolfazl
• Application Number:
  US12/740,806
• Publication Number:
  US 20100296700A1
• Time in Patent Office: 1,669 Days
• Field of Search: 382/100, 382/103, 382/104, 382/236, 340/901, 340/903, 340/905, 340/935, 340/937, 340/938, 340/988, 701/80, 701/65
• US Class Current: 382/104
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• 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 method for detecting the course of a traffic lane, including the following steps:

measuring structures of the traffic lane; evaluating the homogeneity of the measurements; and determining the course of the traffic lane on the basis of the evaluated homogeneity.

Claims

1. Method for detecting a course of a traffic lane, comprising the following steps:

using a processor to perform the steps of;

measuring structures of the traffic lane;

evaluating a homogeneity of the measurements; and

determining the course of the traffic lane on a basis of the evaluated homogeneity,

wherein the homogeneity of the structures is evaluated by determining a density of an area enclosed by the measurements and of a number of measurements.

2. Method according to claim 1, wherein the density is determined as a quotient of the area of a bounding box enclosing the measurements and of the number of measurements.

3. Method according to claim 2, wherein, provided that the measured structures are arranged diagonally, a coordinate system used for the measurements is rotated.

4. Method according to claim 3, wherein the coordinate system is rotated by an iterative method or by principal component analysis.

5. Method according to claim 4, wherein a measure for the area of the bounding box enclosing the measurements is formed from eigenvalues determined by principal component analysis and the area measure formed in this way is divided by the number of measurements in order to obtain the density.

6. Method according to claim 2, wherein the bounding box is determined from differences between the measurements and a prediction of the measurements.

7. Method according to claim 2, wherein the bounding box is determined only from the differences between the measurements and a prediction of the measurements.

8. Method according to claim 1, wherein a beginning and an end of a marking of a dashed traffic lane are determined by detecting transitions on a basis of the measurements.

9. Method according to claim 1, wherein structures of the traffic lane are measured by an image analysis of camera images.

10. Device for detecting a course of a traffic lane, comprising:

a measuring device for measuring structures of the traffic lane; and

an evaluating device for evaluating a homogeneity of the measurements and determining the course of the traffic lane on a basis of the evaluated homogeneity,

wherein the homogeneity of the structures is evaluated by determining a density of an area enclosed by the measurements and of a number of measurements.

11. Device according to claim 10, wherein the device is configured to execute a method.

12. Automatic lane departure warning system or lane keeping system that has a device according to claim 10.

13. Vehicle with an automatic lane departure warning system or lane keeping system according to claim 12.