Detection and classification of light sources using a diffraction grating

• US 9,176,006 B2
• Filed: 01/13/2009
• Issued: 11/03/2015
• Est. Priority Date: 01/15/2008
• Status: Active Grant

• Current Assignee: MobilEye Vision Technologies Ltd. (Intel Corporation)
• Original Assignee: MobilEye Vision Technologies Ltd. (Intel Corporation)
• Inventors: Stein, Gideon
• Primary Examiner(s): RIFKIN, BEN M
• Application Number:
  US12/352,871
• Publication Number:
  US 20090182690A1
• Time in Patent Office: 2,485 Days
• Field of Search: 706/12
• US Class Current: 1/1
• CPC Class Codes: G01J 1/04 Optical or mechanical part supplementary adjustable parts
• CPC Class Codes: G01J 1/0407 Optical elements not provided otherwise, e.g. manifolds, windows, holograms, gratings
• CPC Class Codes: G01J 1/4228 arrangements with two or more detectors, e.g. for sensitivity compensation
• CPC Class Codes: G01J 2001/4247 for testing lamps or other light sources
• CPC Class Codes: G01J 2003/2836 Programming unit, i.e. source and date processing
• CPC Class Codes: G01J 3/02 Details
• CPC Class Codes: G01J 3/0205 Optical elements not provided otherwise, e.g. optical manifolds, diffusers, windows
• CPC Class Codes: G01J 3/0208 using focussing or collimating elements, e.g. lenses or mirrors; performing aberration correction
• CPC Class Codes: G01J 3/18 using diffraction elements, e.g. grating gratings per se G02B
• CPC Class Codes: G01J 3/28 Investigating the spectrum using colour filters G01J3/51
• CPC Class Codes: G01J 3/2803 using photoelectric array detector
• CPC Class Codes: G01J 3/2823 Imaging spectrometer
• CPC Class Codes: G01J 3/46 Measurement of colour; Colour measuring devices, e.g. colorimeters measuring colour temperature G01J5/60
• CPC Class Codes: G02B 5/04 Prisms
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]

Abstract

A system mounted in a vehicle for classifying light sources. The system includes a lens and a spatial image sensor. The lens is adapted to provide an image of a light source on the spatial image sensor. A diffraction grating is disposed between the lens and the light source. The diffraction grating is adapted for providing a spectrum. A processor is configured for classifying the light source as belonging to a class selected from a plurality of classes of light sources expected to be found in the vicinity of the vehicle, wherein the spectrum is used for the classifying of the light source. Both the image and the spectrum may be used for classifying the light source or the spectrum is used for classifying the light source and the image is used for another driver assistance application.

Claims

1. A system mountable in a vehicle, the system for classifying a light source, the system comprising:

a spatial image sensor with a portion that defines a plane;

a lens disposed between the light source and the spatial image sensor, that is adapted to provide an image of the light source on said spatial image sensor;

a diffraction grating or prism disposed between said lens and said light source, that is adapted to direct a spectrum of the light source onto said spatial image sensor in a direction having a vertical component in the plane of said image sensor that tends to position said spectrum at a location on the spatial image sensor that corresponds to sky; and

a processor operatively connectible to said spatial light sensor, wherein said processor is configured to locate said spectrum in the plane of said spatial image sensor and to use said spectrum to classify said light source as belonging to a class selected from a plurality of classes of light sources expected to be found in a vicinity of the vehicle.

2. The system according to claim 1, wherein said spectrum is sensed by said spatial image sensor.

3. The system according to claim 1, wherein the light source is selected from the group consisting of streetlights, headlights, taillights and traffic signs.

4. The system according to claim 1, further comprising a hood enclosure that is configured to limit a field of view so that an image of another light source does not superimpose said spectrum.

5. The system according to claim 1, wherein the processor uses both said image and said spectrum to classify said light source.

6. The system according to claim 1, wherein said image is not used in the classifying, wherein said image is usable for a driver assistance application.

7. The system according to claim 1, wherein said diffraction grating is a holographic diffraction grating.

8. The system according to claim 1, wherein said diffraction grating is a blazed diffraction grating.

9. The system according to claim 1, wherein said spatial image sensor has at least one programmable photosensitive gain region.

10. The system according to claim 1, further comprising a second spatial image sensor that detects said spectrum.

11. The system according to claim 1, further comprising a second spatial image sensor, wherein said spatial image sensor and said second spatial image sensor have different electronic gains.

12. The system according to claim 1, wherein said lens and said diffraction grating are formed in different optical surfaces of a single optical element.

13. The system of claim 1, wherein said image is usable for a driver assistance application and wherein said spectrum is used for classifying the light source by applying rules based on at least one known spectrum of the light sources expected to be found in the road environment.

14. A method of classifying light sources using a system including a camera equipped with a lens and a spatial image sensor, the method comprising:

training at least one classifier based on a plurality of known spectra of light sources expected to be found in a vicinity of the system thereby producing at least one trained classifier;

detecting an image of a light source on the spatial image sensor;

providing a spectrum of the light source at a direction having a vertical component in the plane of said image sensor that tends to position said spectrum at a location on the spatial image sensor that corresponds to sky;

locating said spectrum on the spatial image sensor on the plane of the spatial image sensor; and

classifying the light source based on said spectrum as belonging to a class selected from a plurality of classes of the light sources, by applying said at least one trained classifier.

15. The method, according to claim 14, wherein said classifying is performed by applying rules based on at least one known spectrum of the light sources expected to be found in the vicinity of the system.

16. The method according to claim 14, wherein said at least one classifier includes a least one binary classifier.

17. The method according to claim 14, wherein said at least one classifier is based on support vector machines.

18. The method, according to claim 14, wherein said classifying is based on both said spectrum and said image.

19. The method, according to claim 14, wherein said spectrum is selectably produced by either a diffraction grating or a prism, disposed between said lens and said light source.