Detection of motor vehicle lights with a camera

• US 8,254,632 B2
• Filed: 10/29/2007
• Issued: 08/28/2012
• Est. Priority Date: 11/27/2006
• Status: Active Grant

• Current Assignee: ADC GmbH (TE Connectivity Ltd. (United Kingdom))
• Original Assignee: ADC GmbH (TE Connectivity Ltd. (United Kingdom))
• Inventors: Heinrich, Stefan, Schuler, Reinhard
• Primary Examiner(s): Le, Vu
• Assistant Examiner(s): Abdi, Amara
• Application Number:
  US12/312,391
• Publication Number:
  US 20100061594A1
• Time in Patent Office: 1,765 Days
• Field of Search: 250/208.1, 250/214.A, 349/901, 382/100, 382103-105
• US Class Current: 382/103
• 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 2300/056 Special anti-blinding beams, e.g. a standard beam is chopped or moved in order not to blind
• CPC Class Codes: B60Q 2300/41 preceding vehicle
• CPC Class Codes: B60Q 2300/42 oncoming vehicle
• CPC Class Codes: G06V 20/584 of vehicle lights or traffic lights

Abstract

A method for detecting front headlights and tail lights of a motor vehicle uses a color camera sensor that has a plurality of red pixels, i.e. image points which are only sensitive in the red spectral range, and a plurality of pixels of other colors. In a first evaluation stage, only the intensity of the red pixels in the image is analyzed in order to select relevant points of light in the image.

Claims

1. A method for detecting front headlights and/or tail lights of a motor vehicle with a color camera sensor and a data processing unit, wherein the color camera sensor comprises a plurality of pixels which are sensitive only to red light, and a plurality of pixels which are sensitive to light colors other than red,

characterized in that

in a first evaluation stage for selecting potential motor vehicle headlights or tail lights in the image, only the intensity of those pixels which are sensitive to red light is evaluated, wherein first points of light with an intensity of the red light above a specified first threshold value are identified as potential motor vehicle headlights, and

second points of light with an intensity below a specified second threshold value are identified as potential motor vehicle tail lights.

2. The method according to claim 1, characterized in that the second points of light are further evaluated in that the intensity of the pixels which are only sensitive to red light is compared with the intensity of at least one pixel which is sensitive to a color other than red, and the second points of light further determined by the comparison to be red lights are identified as motor vehicle tail lights.

3. The method according to claim 1, characterized in that the second points of light are further evaluated in that the time intensity progression and/or the absolute intensity and/or the time-dependent movement of the second points of light in the image is observed, in order to differentiate between motor vehicle tail lights and other points of light.

4. The method according to claim 1, characterized in that the first points of light are further evaluated in that the time intensity progression and/or the absolute intensity and/or the time-dependent movement of the first points of light in the image is observed, in order to differentiate between motor vehicle headlights and other points of light.

5. A motor vehicle with a camera sensor and a data processing unit including a computer-readable medium encoded with an image data processing program adapted to execute a method according to claim 1.

6. The motor vehicle according to claim 5, wherein the camera sensor comprises an RGB color camera chip having R pixels, G pixels and B pixels arranged in an alternating two-dimensional array pattern.

7. The method according to claim 1, wherein, as a part of the first evaluation stage, only the intensity of those pixels which are sensitive to red light is evaluated for identifying the second points of light as potential motor vehicle tail lights when this intensity for the second points of light is below the second threshold value.

8. The method according to claim 1, wherein the intensity of all colors of light of the second points of light is evaluated for identifying the second points of light as potential motor vehicle tail lights when this intensity for the second points of light is below the second threshold value.

9. The method according to claim 1, wherein the first threshold value is greater than the second threshold value.

10. The method according to claim 1, wherein the first threshold value is the same as the second threshold value.

11. A method of detecting head lights and tail lights of motor vehicles using a color camera sensor that includes first pixels sensitive to red light and second pixels sensitive to at least one color of light other than red, and a data processing unit, wherein said method comprises the steps:

a) detecting a light source in a field of view of said color camera sensor as a point of light on at least one illuminated first pixel among said first pixels and on at least one illuminated second pixel among said second pixels;

b) in said data processing unit, comparing an output of only said at least one illuminated first pixel to a first threshold, and identifying said light source as a potential head light if said output of only said at least one illuminated first pixel is above said first threshold, and

c) in said data processing unit, comparing an output of said at least one illuminated first pixel and/or said at least one illuminated second pixel to a second threshold, and identifying said light source as a potential tail light if said output of said at least one illuminated first pixel and/or said at least one illuminated second pixel is below said second threshold.

12. The method according to claim 11, wherein said step c) involves comparing said output of only said at least one illuminated first pixel to said second threshold, and identifying said light source as said potential tail light if said output of only said at least one illuminated first pixel is below said second threshold.

13. The method according to claim 12,

wherein said color camera sensor comprises an RGB color camera chip having R pixels as said first pixels, and having G pixels and B pixels as said second pixels,

wherein said R pixels, said G pixels and said B pixels are arranged in an alternating two-dimensional array pattern with said R pixels making up ¼

of a total number of all of said pixels;

wherein said steps b) and c) use only said output of only said R pixels of said RGB color camera chip.

14. The method according to claim 12, further comprising the step:

d) for said light source identified as said potential tail light, in said data processing unit, evaluating said output of said at least one illuminated first pixel and said output of said at least one illuminated second pixel to determine whether said light source is a red light source, and identifying said light source as a tail light if said light source was determined to be a red light source.

15. The method according to claim 14, wherein said evaluating comprises comparing said output of said at least one illuminated first pixel to said output of said at least one second pixel, and determining that said light source is a red light source if said output of said at least one illuminated first pixel is greater than said output of said at least one second pixel.

16. The method according to claim 11, for said light source identified as said potential tail light, further comprising an additional step in said data processing unit, of determining whether said potential tail light is a tail light by evaluating at least one of a time intensity progression, an absolute intensity and a time-dependent movement of said point of light in said field of view of said color camera sensor.

17. The method according to claim 11, for said light source identified as said potential head light, further comprising an additional step in said data processing unit, of determining whether said potential head light is a head light by evaluating at least one of a time intensity progression, an absolute intensity and a time-dependent movement of said point of light in said field of view of said color camera sensor.

18. The method according to claim 11, wherein said first threshold is greater than said second threshold.

19. The method according to claim 11, wherein said first threshold is the same as said second threshold.

20. The method according to claim 11, wherein said detecting of said light source as a point of light in said step a) comprises comparing an output of a selected one of said first pixels to an output of an adjacent one of said first pixels adjacent to said selected one, and determining said point of light as a local maximum if said output of said selected one is greater than said output of said adjacent one.

21. A method of detecting head lights and tail lights of motor vehicles using an RGB color light sensor having R pixels sensitive to red light, G pixels sensitive to green light and B pixels sensitive to blue light arranged in an alternating two-dimensional array pattern, and using an image data processing unit, wherein said method comprises:

a) detecting a point of light on at least one illuminated R pixel and at least one illuminated G pixel or B pixel of said sensor;

b) in said data processing unit, comparing an output intensity of only said at least one illuminated R pixel to a first threshold and to a second threshold;

c) in said data processing unit, identifying said point of light as a potential head light if said output intensity of only said at least one illuminated R pixel is greater than said first threshold;

d) in said data processing unit, identifying said point of light as a potential tail light if said output intensity of only said at least one illuminated R pixel is less than said second threshold;

e) in said data processing unit, if said point of light was identified as a potential head light, then further evaluating an output of at least one of said illuminated pixels to determine whether said point of light is a head light, and

f) in said data processing unit, if said point of light was identified as a potential tail light, then further evaluating an output of at least one of said illuminated pixels to determine whether said point of light is a tail light.