Detection and documentation of tailgating and speeding violations

• US 10,565,867 B2
• Filed: 05/17/2016
• Issued: 02/18/2020
• Est. Priority Date: 05/18/2015
• Status: Active Grant

• Current Assignee: Redflex Traffic Systems Inc.
• Original Assignee: RoadMetric, Ltd. (Redflex Traffic Systems Inc.)
• Inventors: Alon, Yaniv
• Primary Examiner(s): Zhao, Daquan
• Application Number:
  US15/574,159
• Publication Number:
  US 20180137754A1
• Time in Patent Office: 1,372 Days
• Field of Search: 348118, 348116, 348 61, 348 64, 348 73, 348 87, 348 94, 348 95, 348107
• CPC Class Codes: B60Y 2400/3015 Optical cameras
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06T 7/251 involving models
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle
• 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: H04N 23/90 Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums

Abstract

Determining that a trailing vehicle is tailgating a lead vehicle while both are traveling on a lane of a road. A camera captures a timed sequence of image frames at known capture times of the lead vehicle and the trailing vehicle. Images of the trailing vehicle and the lead vehicle are selected from the captured image frames to provide respective image coordinates of the trailing vehicle and the lead vehicle. Multiple world space differential distances between the trailing vehicle and the lead vehicle are computed responsive to the image coordinates in the selected image frames. A tailgating metric responsive to the known capture times of the selected image frames and the computed world space differential distances is determined.

Claims

1. A computerized method for determining that a trailing vehicle is tailgating a lead vehicle while both are traveling on a lane of a road with the trailing vehicle following the lead vehicle, the method comprising:

while a camera is mounted on a moving host vehicle and connected to a processor, capturing from a camera a timed sequence of image frames at known capture times of the lead vehicle and the trailing vehicle;

selecting from the captured image frames selected images of at least a part of the trailing vehicle and at least a part of the lead vehicle, thereby providing respective image coordinates of the trailing vehicle and the lead vehicle in the selected images;

computing responsive to the image coordinates in the selected image frames a plurality of respective world space differential distances between the trailing vehicle and the lead vehicle;

selecting from at least one of the captured image frames a first image point corresponding to an object point on the trailing vehicle and selecting from the at least one captured image frame a second image point corresponding to an object point on the lead vehicle, wherein the first and second image points correspond to world space points defining a line in world space parallel to or included in the plane of the road, wherein the line is parallel to the direction of motion of the trailing vehicle and the lead vehicle, wherein the world space differential distances are computed along said line; and

determining a tailgating metric responsive to the known capture times of the selected image frames and the computed world space differential distances.

2. The computerized method of claim 1, wherein at least one of captured image frames includes the selected images of both the part of the trailing vehicle and the part of the lead vehicle.

3. The computerized method of claim 1, wherein at least one of captured image frames includes the selected image of the part of the trailing vehicle and another of the selected images includes the part of the lead vehicle.

4. The computerized method of claim 1, further comprising:

recognizing from the captured image frames a vehicle model of at least one of the trailing vehicle and the lead vehicle, wherein at least one dimension of the vehicle models is known, wherein said computing the world space differential distances is further responsive to the at least one known dimension of the recognized car model.

5. The computerized method of claim 4 wherein the image frames are captured by a camera with an optical axis substantially perpendicular to the direction of motion of the vehicles and wherein the distance between the trailing vehicle and the lead vehicle is determined relative to the known at least one dimension of the car model.

6. The computerized method of claim 4, wherein the image frames are captured by a camera with an optical axis substantially parallel to the direction of motion of the vehicles and wherein the distance between the trailing vehicle and the lead vehicle is determined relative to the known at least one dimension of the car model.

7. The computerized method of claim 1, wherein at least one of said selected images provides an image of a license plate of the trailing vehicle, and wherein the license number is recognized to associate with the vehicle model and to determine the vehicle model of the trailing vehicle.

8. The computerized method of claim 1, wherein said computing the world space differential distances is further responsive to a range to at least one of the trailing vehicle and the lead vehicle, wherein the range may be determined from the image frames or the range may be measured independently from the image frames using another sensor.

9. The computerized method of claim 1, further comprising:

computing from the image coordinates, world space coordinates (X,Z) of the at least one of the trailing vehicle and the lead vehicle; and

responsive to the known capture times of the selected image frames and the world space coordinates (X,Z), computing a speed of at least one of the trailing vehicle and the lead vehicle.

10. The computerized method of claim 1, wherein the tailgating metric is determined based on the computed world space distances and speed of the trailing vehicle.

11. The computerized method of claim 1, wherein the tailgating metric is determined further based on a time interval between respective capture times of two of the selected captured image frames.

12. The computerized method of claim 1, while the processor is connected to a CAN bus of the host vehicle, the method further comprising:

while the host vehicle is traveling in a lane of the road parallel to the lane of the road on which the lead vehicle and the trailing vehicle are traveling, inputting from the CAN bus the ground speed of the host vehicle; and

from the captured image frames and the known capture times determining the speed of at least one of the trailing vehicle and the lead vehicle relative to the ground speed of the host vehicle.

13. A system including a camera and processor, the system mountable on a moving host vehicle for determining that a trailing vehicle is tailgating a lead vehicle while both are traveling on a lane of a road with the trailing vehicle following the lead vehicle, the system operable to:

capture from a camera a timed sequence of image frames at known capture times of the lead vehicle and the trailing vehicle;

select from the captured image frames selected images of at least a part of the trailing vehicle and at least a part of the lead vehicle, thereby providing respective image coordinates of the trailing vehicle and the lead vehicle in the selected images;

compute responsive to the image coordinates in the selected image frames a plurality of respective world space differential distances between the trailing vehicle and the lead vehicle;

select from at least one of the captured image frames a first image point corresponding to an object point on the trailing vehicle and select from the at least one captured image frame a second image point corresponding to an object point on the load vehicle, wherein the first and second image points correspond to world space points defining a line in world space parallel to or included in the plane of the road, wherein the line is parallel to the direction of motion of the trailing vehicle and the lead vehicle, wherein the world space differential distances are computed along said line; and

determine a tailgating metric responsive to the known capture times of the selected image frames and the computed world space differential distances.

14. The system of claim 13, further configured to recognize from the captured image frames a vehicle model of at least one of the trailing vehicle and the lead vehicle, wherein at least one dimension of the vehicle models is known, wherein the world space differential distance are further responsive to the at least one known dimension of the recognized car model.

15. The system of claim 13, wherein the image frames are captured by a camera with an optical axis substantially perpendicular to the direction of motion of the vehicles and wherein the distance between the trailing vehicle and the lead vehicle is determined relative to the known at least one dimension of the car model.

16. The system of claim 13, wherein the image frames are captured by a camera with an optical axis substantially parallel to the direction of motion of the vehicles and wherein the distance between the trailing vehicle and the lead vehicle is determined relative to the known at least one dimension of the car model.

17. The system of claim 13, wherein at least one of said selected images provides an image of a license plate of the trailing vehicle, and wherein the license number is recognized to associate with and to determine the vehicle model of the trailing vehicle.

18. The system of claim 13, wherein the world space differential distances are responsive to a range to at least one of the trailing vehicle and the lead vehicle, wherein the range may be determined from the image frames or the range may be measured independently from the image frames using another sensor.

19. The system of claim 13, wherein the processor is connectable to a CAN bus of the host vehicle, wherein the system is configured to input from the CAN bus the ground speed of the host vehicle, and from the captured image frames and the known capture times determine the speed of at least one of the trailing vehicle and the lead vehicle relative to the ground speed of the host vehicle.

20. A system mountable on a moving host vehicle for determining that a trailing vehicle is tailgating a lead vehicle while both the trailing vehicle and the lead vehicle are traveling on a lane of a road with the trailing vehicle following the lead vehicle, the system comprising:

multiple cameras mountable on the host vehicle with overlapping fields of view;

a processor configured to capture from the cameras timed sequences of image frames at known capture times of the lead vehicle and the trailing vehicle, wherein the capture of the timed sequences of images frames are synchronized, wherein the processor is further configured to;

select from the captured image frames corresponding image points between the synchronized image frames of at least a part of the trailing vehicle and at least a part of the lead vehicle to provide thereby respective image coordinates of the trailing vehicle and the lead vehicle in at least a portion of the captured image frames;

compute responsive to the image coordinates of the corresponding image points in the synchronized image frames a plurality of world space differential distances between the trailing vehicle and the lead vehicle;

select from at least one of the captured image frames a first image point corresponding to an object point on the trailing vehicle and select from the at least one captured image frame a second image point corresponding to an object point on the lead vehicle, wherein the first and second image points correspond to world space points defining a line in world space parallel to or included in the plane of the road, wherein the line is parallel to the direction of motion of the trailing vehicle and the lead vehicle, wherein the world space differential distances are computed along said line; and

determine a tailgating metric responsive to the known capture times of the selected image frames and the computed world space differential distances.