Collision prediction

Patent metadata

Field	Value
Publication number	US-10402985-B2
Application number	US-201715590258-A
Country	US
Kind code	B2
Filing date	May 9, 2017
Priority date	Sep 21, 2012
Publication date	Sep 3, 2019
Grant date	Sep 3, 2019

Abstract

In some implementations, there is provided a method. The method may include receiving data characterizing a plurality of digital video frames; detecting a plurality of features in each of the plurality of digital video frames; determining, from the detected features, a local scale change and a translational motion of one or more groups of features between at least a pair of the plurality of digital video frames; and calculating a likelihood of collision. Related apparatus, systems, techniques, and articles are also described.

First claim

What is claimed is:

1. A method comprising:
determining, using detected features in a plurality of digital video frames and between at least a pair of the plurality of digital video frames, an inter-frame rate of expansion in an area characterized by at least one group of features;
determining, based on the detected features and between at least the pair of the plurality of digital video frames, an average of an inter-frame difference in pixel position of at least two features in the at least one group of features;
calculating a likelihood of collision, wherein the likelihood of collision includes a ratio between the determined inter-frame rate of expansion and the determined average of inter-frame difference in pixel position;
comparing the likelihood of collision to a threshold; and
generating, in response to the likelihood of collision exceeding the threshold, an alarm that causes a vehicle to turn and/or causes the vehicle to slow down;
wherein at least one of the receiving, the detecting, the determining the inter-frame rate of expansion, the determining the average inter-frame difference in pixel position, the calculating, the comparing, and the generating is performed by at least one data processor.

2. The method of claim 1 , wherein the inter-frame rate of expansion characterizes a difference in an area defined by a convex hull of the at least one group of features.

3. The method of claim 1 , further comprising:
receiving data characterizing the plurality of digital video frames; and
detecting the plurality of features in the plurality of digital video frames.

4. The method of claim 1 , further comprising:
compensating, based on received data, the average inter-frame difference in pixel position, wherein the received data characterizes a motion of a camera, the motion being associated with the plurality of digital video frames.

5. The method of claim 1 , further comprising:
classifying the at least one group of features as being on a collision trajectory by at least comparing a collision point with a predetermined value defining a safety envelope around a camera.

6. A non-transitory computer readable storage medium comprising executable instructions which when executed by at least one processor provides operations comprising:
determining, using detected features in a plurality of digital video frames and between at least a pair of the plurality of digital video frames, an inter-frame rate of expansion in an area characterized by at least one group of features;
determining, based on the detected features and between at least the pair of the plurality of digital video frames, an average of an inter-frame difference in pixel position of at least two features in the at least one group of features;
calculating a likelihood of collision, wherein the likelihood of collision includes a ratio between the determined inter-frame rate of expansion and the determined average inter-frame difference in pixel position;
comparing the likelihood of collision to a threshold; and
generating, in response to the likelihood of collision exceeding the threshold, an alarm that causes a vehicle to turn and/or causes the vehicle to slow down.

7. The non-transitory computer readable storage medium of claim 6 , further comprising:
receiving data characterizing the plurality of digital video frames; and
detecting the plurality of features in the plurality of digital video frames.

8. The non-transitory computer readable storage medium of claim 6 , further comprising:
compensating, based on received data, the average inter-frame difference in pixel position, wherein the received data characterizes a motion of a camera, the motion being associated with the plurality of digital video frames.

9. The non-transitory computer readable storage medium of claim 6 , further comprising:
classifying the at least one group of features as being on a collision trajectory by at least comparing a collision point with a predetermined value defining a safety envelope around a camera.

10. A system comprising:
at least one processor circuitry;
at least one memory storing instructions which, when executed by the at least one processor circuitry, causes operations comprising:
determining, using detected features in a plurality of digital video frames and between at least a pair of the plurality of digital video frames, an inter-frame rate of expansion in an area characterized by at least one group of features;
determining, based on the detected features and between at least the pair of the plurality of digital video frames, an average of an inter-frame difference in pixel position of at least two features in the at least one group of features;
calculating a likelihood of collision, wherein the likelihood of collision includes a ratio between the determined inter-frame rate of expansion and the determined average inter-frame difference in pixel position;
comparing the likelihood of collision to a threshold; and
generating, in response to the likelihood of collision exceeding the threshold, an alarm that causes a vehicle to turn and/or causes the vehicle to slow down.

11. The system of claim 10 , wherein the inter-frame rate of expansion characterizes a difference in an area defined by a convex hull of the at least one group of features.

12. The system of claim 10 , wherein the operations further comprise:
receiving data characterizing the plurality of digital video frames; and
detecting the plurality of features in the plurality of digital video frames.

13. The system of claim 10 , wherein the operations further comprise:
compensating, based on received data, the average inter-frame difference in pixel position, wherein the received data characterizes a motion of a camera, the motion being associated with the plurality of digital video frames.

14. The system of claim 10 , wherein the operations further comprise:
classifying the at least one group of features as being on a collision trajectory by comparing a collision point with a predetermined value defining a safety envelope around a camera.

Assignees

• Schepens Eye Res Inst

Inventors

• Luo Gang
• Peli Eliezer
• Pundlik Shrinivas

Classifications

• B60R2300/8093

  for obstacle warning · CPC title

• G06T2219/2016

  Rotation, translation, scaling · CPC title

• G06T2207/30248

  Vehicle exterior or interior · CPC title

• G08G1/16

  Anti-collision systems (road vehicle drive control systems for predicting or avoiding probable or impending collision otherwise than by control of a particular sub-unit B60W30/08) · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title