TRAFFIC VIOLATION DETECTION
First Claim
1. A method to detect a traffic violation, the method comprising:
- acquiring a sequence of image frames from at least one image source recording vehicle traffic;
defining a first region of interest (ROI) on an image plane of at least one of the image frames in the sequence of image frames;
searching the defined first ROI in at least one of the image frames in the sequence of image frames for a candidate violating vehicle that may be committing the traffic violation;
in response to detecting the candidate violating vehicle in the defined first ROI in at least one image frame in the sequence of image frames, tracking a location of the detected candidate violating vehicle in at least one other of the image frames in the sequence of image frames;
generating at least one trajectory of the detected candidate violating vehicle using the location of the detected candidate violating vehicle in the at least one image frame and in the at least one other of the image frames;
defining a second ROI on an image plane of at least one of the image frames in the sequence of image frames; and
in response to determining the at least one trajectory of the detected candidate violating vehicle includes a trajectory component in the defined first ROI and the defined second ROI, classifying the candidate violating vehicle as a violating vehicle.
6 Assignments
0 Petitions
Accused Products
Abstract
A method for detecting a vehicle running a stop signal positioned at an intersection includes acquiring a sequence of frames from at least one video camera monitoring an intersection being signaled by the stop signal. The method includes defining a first region of interest (ROI) including a road region located before the intersection on the image plane. The method includes searching the first ROI for a candidate violating vehicle. In response to detecting the candidate violating vehicle, the method includes tracking at least one trajectory of the detected candidate violating vehicle across a number of frames. The method includes classifying the candidate violating vehicle as belonging to one of a violating vehicle and a non-violating vehicle based on the at least one trajectory.
-
Citations
24 Claims
-
1. A method to detect a traffic violation, the method comprising:
-
acquiring a sequence of image frames from at least one image source recording vehicle traffic; defining a first region of interest (ROI) on an image plane of at least one of the image frames in the sequence of image frames; searching the defined first ROI in at least one of the image frames in the sequence of image frames for a candidate violating vehicle that may be committing the traffic violation; in response to detecting the candidate violating vehicle in the defined first ROI in at least one image frame in the sequence of image frames, tracking a location of the detected candidate violating vehicle in at least one other of the image frames in the sequence of image frames; generating at least one trajectory of the detected candidate violating vehicle using the location of the detected candidate violating vehicle in the at least one image frame and in the at least one other of the image frames; defining a second ROI on an image plane of at least one of the image frames in the sequence of image frames; and in response to determining the at least one trajectory of the detected candidate violating vehicle includes a trajectory component in the defined first ROI and the defined second ROI, classifying the candidate violating vehicle as a violating vehicle. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13)
-
-
14. A system to detect a traffic violation, the system including a memory and a processor in communication with the memory, the processor configured to:
-
acquire a sequence of image frames from at least one image source recording vehicle traffic; define a first region of interest (ROI) on an image plane of at least one of the image frames in the sequence of image frames; search the defined first ROI in at least one of the image frames in the sequence of image frames for a candidate violating vehicle that may be committing the traffic violation; in response to detecting the candidate violating vehicle in the defined first ROI in at least one image frame in the sequence of image frames, track a location of the detected candidate violating vehicle in at least one other of the image frames in the sequence of image frames; generate at least one trajectory of the detected candidate violating vehicle using the location of the detected candidate violating vehicle in the at least one image frame and in the at least one other of the image frames; define a second ROI on an image plane of at least one of the image frames in the sequence of image frames; and classify the detected candidate violating vehicle as a violating vehicle if the at least one trajectory of the detected candidate violating vehicle includes a trajectory component in the defined first ROI and the defined second ROI. - View Dependent Claims (15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A method to detect a movement violation, the method comprising:
-
acquiring a sequence of image frames from at least one image source recording object movement; defining a first region of interest (ROI) on an image plane of at least one of the image frames in the sequence of image frames; searching the defined first ROI in at least one of the image frames in the sequence of image frames for a candidate violating object that may be committing the movement violation; in response to detecting the candidate violating object in the defined first ROI in at least one image frame in the sequence of image frames, tracking a location of the detected candidate violating object in at least one other of the image frames in the sequence of image frames; generating at least one trajectory of the detected candidate violating object using the location of the detected candidate violating object in the at least one image frame and in the at least one other of the image frames; defining a second ROI on an image plane of at least one of the image frames in the sequence of image frames; and classifying the detected candidate violating object as a violating object if the at least one trajectory of the detected candidate violating object includes a trajectory component in the defined first ROI and the defined second ROI.
-
Specification