Detection and categorisation of points of light with a camera in a motor vehicle environment

• US 8,218,009 B2
• Filed: 09/27/2007
• Issued: 07/10/2012
• 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, Fechner, Thomas, Almeida, Carlos
• Primary Examiner(s): Etienne, Ario
• Assistant Examiner(s): Williams, Clayton R
• Application Number:
  US12/311,634
• Publication Number:
  US 20100026806A1
• Time in Patent Office: 1,748 Days
• Field of Search: None
• US Class Current: 348/148
• CPC Class Codes: B60Q 1/1423 Automatic dimming circuits, i.e. switching between high beam and low beam due to change of ambient light or light level in road traffic
• 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
• 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 and categorizing points of light for a motor vehicle with a camera sensor directed towards the motor vehicle environment is presented. Here, at least one first category for passive, illumined reflectors and at least one second category for self-radiating, moving lights, in particular motor vehicle lights, is provided. For this purpose, the time progression of the intensity of a point of light is analysed. On the basis of the intensity fluctuation, points of light are categorized as motor vehicle lights or as reflectors.

Claims

1. A method for detecting and categorizing points of light for a subject motor vehicle with a camera sensor directed toward the motor vehicle environment around the subject motor vehicle, which provides at least one first category for stationary, illuminated reflectors used to mark an edge of a roadway, and at least one second category for self-radiating, moving, motor vehicle lights of other motor vehicles, comprising the steps:

recording an image sequence of the motor vehicle environment;

tracking at least one point of light in the image sequence;

determining a varying intensity of the point of light in at least two images of the image sequence, and analyzing a time progression of the intensity as an intensity fluctuation, and

categorizing a point of light with a comparatively high intensity fluctuation as a motor vehicle light, and categorizing a point of light with a comparatively low intensity fluctuation as a reflector.

2. The method for detecting and categorizing points of light according to claim 1, wherein, in order to analyze the intensity fluctuation of a point of light, the variance or standard deviation is calculated with reference to a continuous average value.

3. The method for detecting and categorizing points of light according to claim 1, wherein, in order to analyze the intensity fluctuation of a point of light, the swing range of consecutive measured values is calculated.

4. The method for detecting and categorizing points of light according to claim 1, further comprising analyzing an intensity of points of light in the image.

5. The method for detecting and categorizing points of light according to claim 4, wherein points of light with the intensity above a first threshold value are detected as the motor vehicle lights, and points of light with the intensity below a second threshold value are detected as the reflectors.

6. The method for detecting and categorizing points of light according to claim 1, further comprising analyzing a movement of the points of light in the image sequence, and taking into account the subject motor vehicle'"'"'s own movement and/or a distance estimation of a selected point of light, the selected point of light is classified as being stationary or moving.

7. The method for detecting and categorizing points of light according to claim 1, further comprising analyzing a pitch angle of the subject motor vehicle and, with a movement analysis of the points of light, extracting an interference caused by a pitch movement of the subject motor vehicle.

8. A motor vehicle with a camera sensor and image data processing program implementing a method according to claim 1.

9. A motor vehicle with a camera sensor for a passenger assistance system, for recording a driving lane, and an image data processing program implementing a method according to claim 1.

10. The method for detecting and categorizing points of light according to claim 1, further comprising switching an illumination condition of headlights of the subject motor vehicle in response to and dependent on whether the point of light is categorized as the motor vehicle light or as the reflector.

11. A method of controlling a motor vehicle system of a subject motor vehicle that is equipped with a camera sensor and an image processor, wherein said camera sensor is arranged to monitor an environment outside of said motor vehicle, said image processor is connected and adapted to evaluate image data provided by said camera sensor, and said method comprises:

a) with said camera sensor, acquiring an image sequence of successive images of said environment;

b) in said image processor, respectively detecting and tracking at least one respective point of light in said successive images in said image sequence;

c) in said image processor, determining a respective luminous intensity of said respective point of light, and determining a respective intensity fluctuation of said respective point of light as a fluctuation of said respective luminous intensity among said successive images in said image sequence;

d) in said image processor, evaluating said respective intensity fluctuation as to whether said respective intensity fluctuation falls into a first category of comparatively high intensity fluctuations or a second category of comparatively low intensity fluctuations compared to said first category;

e) in said image processor, in response to and dependent on at least a result of said evaluating, respectively identifying said respective point of light as a motor vehicle light of another motor vehicle in said environment if said respective intensity fluctuation was evaluated as falling into said first category of comparatively high intensity fluctuations, and respectively identifying said respective point of light as a stationary reflector in said environment if said respective intensity fluctuation was evaluated as falling into said second category of comparatively low intensity fluctuations; and

f) in a controller of said subject motor vehicle, automatically controlling a motor vehicle system of said subject motor vehicle in response to and dependent on whether said respective point of light was identified as said motor vehicle light or as said stationary reflector.

12. The method according to claim 11, wherein said motor vehicle system comprises a lighting system including headlights of said subject motor vehicle, and said controlling comprises switching an illumination condition of said headlights.

13. The method according to claim 11, wherein said determining of said respective intensity fluctuation comprises comparing respective individual values of said respective luminous intensity, which were respectively determined individually in said successive images, to an average of said respective luminous intensity among said successive images to determine a variance or standard deviation of said individual values relative to said average.

14. The method according to claim 13, further comprising determining said average as a continuous moving average of successive ones of said individual values of said respective luminous intensity among said successive images.

15. The method according to claim 11, wherein said determining of said respective intensity fluctuation comprises determining a difference range of consecutive individual values of said respective luminous intensity, which were respectively determined individually in consecutive ones of said successive images.

16. The method according to claim 11, wherein said evaluating comprises comparing said respective intensity fluctuation to at least one threshold.

17. The method according to claim 16, further comprising determining said at least one threshold respectively as a function of intensity fluctuation relative to luminous intensity, ranging from higher values of intensity fluctuation at lower values of luminous intensity, to lower values of intensity fluctuation at higher values of luminous intensity.

18. The method according to claim 17, wherein said function is a linear function with a negative slope of intensity fluctuation relative to luminous intensity.

19. The method according to claim 16, further comprising establishing said at least one threshold to include a first threshold below most arising values of intensity fluctuation of all motor vehicle lights in said image sequence, and a second threshold above most arising values of intensity fluctuation of all stationary reflectors in said image sequence.

20. The method according to claim 19, wherein said first threshold and said second threshold are both defined by a single same threshold function.

21. The method according to claim 19, wherein said establishing of said first threshold and said second threshold comprises evaluating and categorizing past values of intensity fluctuation of prior points of light that were previously imaged.

22. The method according to claim 11,

further comprising, in said image processor, evaluating said respective luminous intensity with regard to a distance-dependent variation thereof, and ascertaining whether said distance-dependent variation more closely matches a first function 1/x² or a second function 1/x⁴ wherein x is a distance from said camera sensor to an object in said environment appearing as said respective point of light in said successive images in said image sequence, and

wherein said identifying of said respective point of light is further in response to and dependent on a result of said ascertaining, wherein said respective point of light is identified as said motor vehicle light if said distance-dependent variation more closely matches said first function 1/x² and said respective point of light is identified as said stationary reflector if said distance-dependent variation more closely matches said second function 1/x^4.