VEHICLE IMAGING SYSTEM AND METHOD FOR CATEGORIZING OBJECTS USING RELATIVE MOTION ANALYSIS

• US 20140247351A1
• Filed: 02/20/2014
• Published: 09/04/2014
• Est. Priority Date: 03/01/2013
• Status: Active Grant

• Current Assignee: HL Klemove Corporation
• Original Assignee: Gentex Corporation
• Inventors: Falb, David M., Murillo Amaya, Sergio E.
• Granted Patent: US 9,185,363 B2
• Time in Patent Office: Days
• US Class Current: 348/148
• CPC Class Codes: B60Q 1/085 due to special conditions, e.g. adverse weather, type of road, badly illuminated road signs or potential dangers B60Q1/10, B60Q1/12, B60Q1/1423 take precedence
• 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
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

An imaging system is provided for a vehicle. The imaging system includes: an imager configured to image a forward external scene of the controlled vehicle and to generate image frame data corresponding to each frame of a series of acquired image frames; and a processor configured to receive and analyze the image frame data to detect a dominant scene motion and to determine relative motion of objects as compared to the dominant scene motion, and wherein the analysis of the image frame data performed by the processor includes categorizing objects detected in the image frame data using the relative motion of those objects.

Claims

1. An imaging system for a vehicle, comprising:

an imager configured to image a scene external and forward of the vehicle and to generate image frame data corresponding to each frame of a series of acquired image frames; and

a processor configured to receive and analyze the image frame data to detect a dominant scene motion and to determine relative motion of objects relative to the dominant scene motion,

wherein the analysis of the image frame data performed by said processor includes categorizing objects detected in the image frame data using the relative motion of those objects.

2. The imaging system of claim 1, wherein said processor detects a dominant scene motion based upon objects detected in a small central window within the image frames.

3. The imaging system of claim 1, wherein said processor determines the relative motion of objects relative to the dominant scene motion in part by determining the movement of each of the objects throughout the series of acquired image frames and then comparing such movement to the dominant scene motion.

4. The imaging system of claim 3, wherein said processor generates a light list based upon the analysis of the image frame data, the light list identifying objects in an image frame along with positions of the objects, and wherein said processor maintains a light list history for the series of acquired image frames whereby movement of each of the objects throughout is determined by comparing positions of each of the objects in each light list.

5. The imaging system of claim 4, wherein said processor determines relative movement for those objects in a light list for an image frame that were identified in at least three prior light lists.

6. The imaging system of claim 1, wherein said processor categorizes objects detected in the image frame data using one or more of a detected brightness, size, position of center of mass, age, and color of the objects in addition to the relative motion of those objects.

7. The imaging system of claim 6, wherein, when categorizing objects, said processor increases the probability that an object is a vehicle if the object has a high relative motion.

8. An exterior light control system for controlling exterior lights of a controlled vehicle, comprising the imaging system of claim 1, wherein said processor generates an exterior light control signal that is used to control the exterior lights in response to analysis of the image data.

9. A method for categorizing objects detected in a scene forward of a vehicle, the method comprising:

imaging a scene external and forward of the controlled vehicle and generating image frame data corresponding to each frame of a series of acquired image frames; and

receiving and analyzing the image frame data in a processor to detect a dominant scene motion, to determine relative motion of objects relative to the dominant scene motion, and to categorize objects detected in the image frame data using the relative motion of those objects.

10. The method of claim 9, wherein the relative motion of objects is determined relative to the dominant scene motion in part by determining the movement of each of the objects throughout the series of acquired image frames and then comparing such movement to the dominant scene motion.

11. The method of claim 10, wherein a light list is generated that identifies objects in an image frame along with positions of the objects, and wherein a light list history is maintained for the series of acquired image frames whereby movement of each of the objects throughout is determined by comparing positions of each of the objects in each light list.

12. The method of claim 11, wherein relative movement is determined for those objects in a light list for an image frame that were identified in at least three prior light lists.

13. The method of claim 9, wherein objects detected in the image frame data are categorized using one or more of a detected brightness, size, position of center of mass, age, and color of the objects in addition to the relative motion of those objects.

14. The method of claim 13, wherein, when categorizing objects, the probability that an object is a vehicle is increased if the object has a high relative motion.

15. A non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor, cause the processor to categorize objects in a scene forward of a vehicle, by executing the steps comprising:

imaging a scene external and forward of the vehicle and generating image data corresponding to each frame of a series of acquired image frames; and

receiving and analyzing the image frame data in a processor to detect a dominant scene motion, to determine relative motion of objects relative to the dominant scene motion, and to categorize objects detected in the image frame data using the relative motion of those objects.

16. The non-transitory computer readable medium of claim 15, wherein the software instructions further cause the processor to generate a control signal that is used to control exterior lights of the vehicle in response to analysis of the image data.

17. The non-transitory computer readable medium of claim 16, wherein the relative motion of objects is determined relative to the dominant scene motion in part by determining the movement of each of the objects throughout the series of acquired image frames and then comparing such movement to the dominant scene motion.

18. The non-transitory computer readable medium of claim 17, wherein a light list is generated that identifies objects in an image frame along with positions of the objects, and wherein a light list history is maintained for the series of acquired image frames whereby movement of each of the objects throughout is determined by comparing positions of each of the objects in each light list.

19. The non-transitory computer readable medium of claim 18, wherein relative movement is determined for those objects in a light list for an image frame that were identified in at least three prior light lists.

20. The non-transitory computer readable medium of claim 15, wherein objects detected in the image frame data are categorized using one or more of a detected brightness, size, position of center of mass, age, and color of the objects in addition to the relative motion of those objects.