Method for distinguishing between reflectors arranged on the roadside and vehicle lights in the dark

• US 8,587,456 B2
• Filed: 10/15/2008
• Issued: 11/19/2013
• Est. Priority Date: 10/25/2007
• Status: Expired due to Fees

• Current Assignee: ADC GmbH (TE Connectivity Ltd. (United Kingdom))
• Original Assignee: ADC GmbH (TE Connectivity Ltd. (United Kingdom))
• Inventors: Walter, Michael, Fechner, Thomas
• Primary Examiner(s): Lee, Benjamin C
• Assistant Examiner(s): BLACK-CHILDRESS, RAJSHEED O
• Application Number:
  US12/739,760
• Publication Number:
  US 20100295705A1
• Time in Patent Office: 1,861 Days
• Field of Search: None
• US Class Current: 340/937
• CPC Class Codes: B60Q 1/143 combined with another condition, e.g. using vehicle recognition from camera images or activation of wipers
• CPC Class Codes: B60Q 2300/322 Road curvature
• CPC Class Codes: B60Q 2300/41 preceding vehicle
• CPC Class Codes: B60Q 2300/42 oncoming vehicle
• CPC Class Codes: B60Q 2300/45 Special conditions, e.g. pedestrians, road signs or potential dangers
• CPC Class Codes: G06V 20/584 of vehicle lights or traffic lights

Abstract

A method and a device for distinguishing between reflectors arranged on the roadside and vehicle lights and for identifying said reflectors and vehicle lights for use in an automatic light control system in a vehicle with a camera sensor oriented towards the surroundings of the vehicle. The method includes recording a sequence of images of the surroundings of the vehicle. Light spots are found and subjected to temporal tracking within the sequence of images. Afterwards, for each light spot, a decision is made whether the light source of the light spot is a vehicle light or a reflector. A statistical map for the entire camera image is determined from the position of the light spots found with the statistical map comprising regions that are predominantly occupied by reflectors and regions that are predominantly occupied by vehicles driving ahead or by oncoming vehicles. All light sources that are found within the regions occupied by reflectors are identified as potential reflectors and all light sources within the regions occupied by vehicle lights are identified as potential vehicle lights.

Claims

1. A method for distinguishing between reflectors (R) arranged on the roadside and vehicle lights (FL1, FL2) and identifying said reflectors (R) and vehicle lights (FL1, FL2) for use in a vehicle with a camera sensor oriented towards the surroundings of the vehicle, comprising:

recording a sequence of images of the surroundings of the vehicle;

finding and temporally tracking light spots within the sequence of images;

deciding, for each light spot, whether the light source of the light spot is a vehicle light (FL1, FL2) or a reflector (R);

determining a statistical map for the entire camera image from the position of the light spots found, by defining reflector regions within the camera image that are predominantly occupied by reflectors (R) and defining vehicle regions within the camera image that are predominantly occupied by vehicles driving ahead (FL2) or by oncoming vehicles (FL1), the reflector regions being separate from the vehicle regions;

identifying all light sources that are found within the reflector regions as potential reflectors (R); and

identifying all light sources within the vehicle regions as potential vehicle lights (FL1, FL2).

2. The method according to claim 1, wherein light spots that are arranged, in the image, in line with further light spots are identified as potential reflectors (R).

3. The method according to claim 1, wherein light spots whose spatial arrangement corresponds to a periodic arrangement of light sources in the real space are identified as potential reflectors (R).

4. The method according to claim 1, wherein a statistical map in world coordinates is created by taking into consideration the camera fitting position, said statistical map in world coordinates comprising regions that are predominantly occupied by reflectors (R) and regions that are predominantly occupied by vehicles driving ahead or by oncoming vehicles (FL1, FL2).

5. The method according to claim 1, wherein deciding, for each light spot, whether the light source of the light spot is a vehicle light (FL1, FL2) or a reflector (R) further comprises considering the direction and the length of a displacement vector for the tracked light spot.

6. The method according to claim 1, wherein the point in time of the decision whether the light source is a vehicle (FL1, FL2) or a reflector (R) is delayed or accelerated in dependence on the position where the light spot is found and on the intensity of the light spot.

7. The method according to claim 1, wherein the steering angle and/or the yaw angle are/is known and the curvature of the map is adapted to the course of the road.

8. The method according to claim 1, wherein a sensor-based lane detection system is provided that supplies information about the lane lying ahead so that the position of the map regions in which potential reflectors (R) can be found is adapted according to the information about the roadway curvature lying ahead.

9. The method according to claim 8, wherein the statistical map is adapted to the roadway width determined by the lane detection system.

10. The method according to claim 8, wherein the statistical map is adapted to the current driving behaviour determined by the lane detection system.

11. A device for distinguishing between and for identifying reflectors (R) arranged on the roadside and vehicle lights (FL1, FL2) in the surroundings of a vehicle, said device comprising a camera sensor and an evaluation unit, wherein the evaluation unit evaluates the image data acquired by the camera sensor and applies, for this purpose, a method according to claim 1.

12. The method according to claim 1, wherein identification of all light sources that are found outside the vehicle regions and reflector regions is delayed until further features of the light sources allow identification to be performed.