SYSTEM AND METHOD FOR PREDICTING VEHICLE LOCATION

• US 20130100286A1
• Filed: 10/21/2011
• Published: 04/25/2013
• Est. Priority Date: 10/21/2011
• Status: Active Grant

• Current Assignee: Mesa Engineering, Inc.
• Original Assignee: Mesa Engineering, Inc.
• Inventors: LAO, Yuanwei
• Granted Patent: US 8,970,701 B2
• Time in Patent Office: Days
• US Class Current: 348/148
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G08G 1/0175 by photographing vehicles, e.g. when violating traffic rules
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors
• CPC Class Codes: G08G 1/054 photographing overspeeding vehicles

Abstract

A system and method for detecting a traffic signal violation. In one embodiment a system includes a video image capture device, an image rectifier, and a vehicle position predictor. The image rectifier is configured to provide a rectified image by applying a rectification function an image captured by the video image capture device. The vehicle position predictor is configured to provide a prediction, based on the rectified image, of whether a vehicle approaching an intersection will enter the intersection.

Claims

1. A traffic signal violation detection system, comprising:

a video image capture device;

an image rectifier configured to provide a rectified image by applying a rectification function to an image captured by the video image capture device; and

a vehicle position predictor configured to provide a prediction, based on the rectified image, of whether a vehicle approaching an intersection will enter the intersection.

2. The system of claim 1, further comprising a battery;

wherein the system is self-contained and powered by the battery.

3. The system of claim 1, further comprising a vehicle tracker configured to track vehicle position over a monitored path to a horizon of the video image capture device field of view.

4. The system of claim 1, wherein the image rectifier is configured to determine a vanishing point as an intersection of parallel lines at a horizon from one or more images captured by the video image capture device and generate the rectified image based on the vanishing point.

5. The system of claim 4, wherein the image rectifier is configured to determine the vanishing point by projecting lines along edges of detected vehicles to a point where the edges intersect.

6. The system of claim 4, wherein the image rectifier is configured to determine the vanishing point by tracking edges of vehicles over a plurality of images to a point where the edges intersect.

7. The system of claim 4, wherein the image rectifier is configured to determine the vanishing point by tracking vehicles over a plurality of images to a point where the vehicles intersect.

8. The system of claim 1, wherein the vehicle position predictor is configured to determine speed of the vehicle approaching the intersection based on the rectified image.

9. The system of claim 1, wherein the vehicle position predictor is configured to predict whether the vehicle approaching the intersection will enter the intersection based on vehicle speed determined based on a plurality of rectified images of the vehicle approaching the intersection.

10. The system of claim 1, further comprising a still image capture device coupled to the vehicle position predictor;

wherein the vehicle position predictor is configured to trigger the still image capture device to capture an image of the vehicle in the intersection based on the prediction.

11. The system of claim 1, further comprising an optical signal state detector configured to determine a state of a traffic signal at the intersection based on light emitted by the traffic signal.

12. The system of claim 1, further comprising a line generator configured to determine an intersection boundary line based on a stopping position of a plurality of vehicles at the intersection.

13. A method for predicting signal violations, comprising:

capturing, by an intersection monitoring device, a video image of a street intersection;

identifying, by the intersection monitoring device, a vehicle approaching the intersection in a frame of the video image;

rectifying, by the intersection monitoring device, a frame of the video image; and

predicting, by the intersection monitoring device, based on the rectified image, whether the vehicle approaching the intersection will enter the intersection without stopping.

14. The method of claim 13, further comprising tracking trajectories of a plurality of vehicles over a monitored path to a horizon of the monitoring device field of view.

15. The method of claim 13, further comprising:

determining a first vanishing point as an intersection of parallel lines extending in a direction of monitored traffic flow from one or more images captured by the monitoring device;

determining a second vanishing point as a intersection of parallel lines extending perpendicular to the direction of monitored traffic flow from one or more images captured by the monitoring device; and

generating the rectified image based on the vanishing points.

16. The method of claim 15, further comprising determining at least one of the first and second vanishing points by projecting lines along edges of detected vehicles to a point where the edges intersect.

17. The method of claim 15, further comprising determining the first vanishing point by tracking edges of vehicles over a plurality of images to a point where the edges intersect.

18. The method of claim 15, further comprising determining the first vanishing point by tracking vehicles over a plurality of images to a point where the vehicles intersect.

19. The method of claim 13, further comprising estimating speed of the vehicle approaching the intersection based on the rectified image.

20. The method of claim 13, wherein the predicting comprises determining whether the vehicle approaching the intersection will enter the intersection based on an estimated speed of the vehicle determined based on a plurality of rectified image of the vehicle approaching the intersection.

21. The method of claim 13, further comprising:

triggering a still image capture device based on a result of the predicting; and

capturing a still image of the vehicle in the intersection.

22. The method of claim 21, further comprising:

determining a state of a traffic signal at the intersection using a light color sensor focused on the traffic signal;

wherein the triggering is responsive to determining the light color to be red.

23. The method of claim 13, further comprising determining an intersection boundary line based on the positions of a plurality of vehicles stopped at the intersection.

24. A computer-readable storage medium encoded with instructions that when executed cause a processor to:

store a video image of a street intersection;

identify a vehicle approaching the intersection in a frame of the video image;

apply a rectification function to the video image to produce a rectified image; and

predict, based on the rectified image, whether the vehicle approaching the intersection will enter the intersection.

25. The computer-readable medium of claim 24, further comprising instructions that when executed cause the processor to track trajectories of a plurality of vehicles over a monitored path to a horizon.

26. The computer-readable medium of claim 24, further comprising instructions that when executed cause the processor to:

determine a vanishing point as an intersection of parallel lines at a horizon from the video image; and

generate the rectified image based on the vanishing point.

27. The computer-readable medium of claim 26, further comprising instructions that when executed cause the processor to determine the vanishing point by at least one of:

projecting lines along edges of detected vehicles to a point where the edges intersect;

tracking edges of vehicles over a plurality of images to a point where the edges intersect; and

tracking vehicles over a plurality of images to a point where the vehicles intersect.

28. The computer-readable medium of claim 24, further comprising instructions that when executed cause the processor to:

estimate speed of the vehicle approaching the intersection based on the rectified image; and

determine whether the vehicle approaching the intersection will enter the intersection based on an estimated speed of the vehicle at a predetermined distance from the intersection.

29. The computer-readable medium of claim 24, further comprising instructions that when executed cause the processor to trigger a still image capture device to capture a still image of the vehicle in the intersection based on a result of the prediction.

30. The computer-readable medium of claim 28, further comprising instructions that when executed cause the processor to:

determine a state of a traffic signal at the intersection using a light color sensor focused on the traffic signal; and

trigger the still image capture device responsive to determining the light color to indicate a stop state.

31. The computer-readable medium of claim 28, further comprising instructions that when executed cause the processor to determine an intersection boundary line based the positions of a plurality of vehicles stopped at the intersection.