Systems and methods for detecting vehicle tailgating
A device may obtain video data associated with a driving event involving a first vehicle. The device may determine a vanishing point associated with the video data and may construct a cone of impact of the first vehicle based on the vanishing point. The device may detect a second vehicle within the cone of impact and may analyze the subset of video frames to determine a distance between the first vehicle and the second vehicle. The device may determine a speed of the first vehicle during a time period associated with a subset of video frames. The device may determine a headway score, representative of a severity associated with the first vehicle being within a proximity threshold of the second vehicle during the time period, based on the distance and the speed. The device may determine an occurrence of a tailgating event based on the headway score.
1. A method, comprising:
determining, by a device, that a first vehicle is within a cone of impact in a video frame based on determining whether a center point of a lowest edge of a bounding box associated with the first vehicle is within the cone of impact,
wherein the cone of impact is determined from a perspective of a camera that captures the video frame and is represented as a triangular area of the video frame,
wherein a vertex of the triangular area includes a vanishing point associated with the video frame,
wherein a side of the triangular area coincides with a portion of a bottom boundary of the video frame, and
wherein the portion of the bottom boundary of the video frame is selected based on a horizontal inclination of the video frame and a height at which the camera is mounted;
determining, by the device, a distance between the first vehicle and a second vehicle based on the cone of impact;
determining, by the device, a score based on the distance between the first vehicle and the second vehicle and a speed of the second vehicle; and
providing, by the device and to a client device, a notification based on the score.
2. The method of claim 1 , wherein the score represents a severity of tailgating event associated with the first vehicle and the second vehicle.
3. The method of claim 1 , further comprising:
utilizing a machine learning model to detect the first vehicle within the cone of impact in the video frame.
4. The method of claim 1 , further comprising:
receiving authentication of the score, and
training a tailgate detection model based on the score.
5. The method of claim 1 , further comprising:
utilizing an object detection algorithm to identify the first vehicle within the video frame.
6. The method of claim 1 , wherein the vanishing point is determined based on selecting a candidate point, from a set of candidate points, based on confidence scores associated with each candidate point of the set of candidate points.
7. The method of claim 1 , wherein a length of the portion of the bottom boundary is determined based on one or more of:
the height at which the camera is mounted, and
a size of the first vehicle in the video frame.
8. A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:
one or more instructions that, when executed by one or more processors of a device, cause the device to:
determine that a first vehicle is within a cone of impact in a video frame based on determining whether a center point of an edge of a bounding box associated with the first vehicle is within the cone of impact,
wherein the cone of impact is determined from a perspective of a camera that captures the video frame and is represented as a triangular area of the video frame,
wherein a vertex of the triangular area includes a vanishing point associated with the video frame,
wherein a side of the triangular area coincides with a portion of a bottom boundary of the video frame, and
wherein the portion of the bottom boundary of the video frame is selected based on a horizontal inclination of the video frame and a height at which the camera is mounted;
determine a distance between the first vehicle and a second vehicle based on the cone of impact;
determine a score based on the distance between the first vehicle and the second vehicle and a speed of the second vehicle; and
provide, to a client device, a notification based on the score.
9. The non-transitory computer-readable medium of claim 8 , wherein the score represents a severity of tailgating event associated with the first vehicle and the second vehicle.
10. The non-transitory computer-readable medium of claim 8 , wherein the one or more instructions further cause the device to:
utilize a machine learning model to detect the first vehicle within the cone of impact in the video frame.
11. The non-transitory computer-readable medium of claim 8 , wherein the one or more instructions further cause the device to:
receive authentication of the score, and
train a tailgate detection model based on the score.
12. The non-transitory computer-readable medium of claim 8 , wherein the one or more instructions further cause the device to:
utilize an object detection algorithm to identify the first vehicle within the video frame.
13. The non-transitory computer-readable medium of claim 8 , wherein the vanishing point is determined based on selecting a candidate point, from a set of candidate points, based on confidence scores associated with each candidate point of the set of candidate points.
14. The non-transitory computer-readable medium of claim 8 , wherein a length, of the side of the triangular area that coincides with the portion of the bottom boundary of the video frame, is determined based on:
the height at which the camera is mounted, and
a size of the first vehicle in the video frame.
15. A device, comprising:
one or more processors configured to:
determine that a first vehicle is within a cone of impact in a video frame based on determining whether a center point associated with an edge of a bounding box associated with the first vehicle is within the cone of impact,
wherein the cone of impact is determined from a perspective of a camera that captures the video frame and is represented as a triangular area of the video frame,
wherein a vertex of the triangular area includes a vanishing point associated with the video frame,
wherein a side of the triangular area coincides with a portion of a bottom boundary of the video frame, and
wherein the portion of the bottom boundary of the video frame is selected based on a horizontal inclination of the video frame and a height at which the camera is mounted on a second vehicle;
determine a score based on a distance between the first vehicle and the second vehicle and a speed of the second vehicle; and
provide, to a client device, a notification based on the score.
16. The device of claim 15 , wherein the score represents a severity of tailgating event associated with the first vehicle and the second vehicle.
17. The device of claim 15 , wherein the one or more processors are further configured to:
utilize a machine learning model to detect the first vehicle within the cone of impact in the video frame.
18. The device of claim 15 , wherein the one or more processors are further configured to:
receive authentication of the score, and
train a tailgate detection model based on the score.
19. The device of claim 15 , wherein the one or more processors are further configured to:
utilize an object detection algorithm to identify the first vehicle within the video frame.
20. The device of claim 15 , wherein the vanishing point is determined based on selecting a candidate point, from a set of candidate points, based on confidence scores associated with each candidate point of the set of candidate points.