Event detection in video surveillance

Patent metadata

Field	Value
Publication number	US-12243304-B2
Application number	US-202217683820-A
Country	US
Kind code	B2
Filing date	Mar 1, 2022
Priority date	Mar 31, 2021
Publication date	Mar 4, 2025
Grant date	Mar 4, 2025

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for retroactive event detection. The methods, systems, and apparatus include actions of obtaining an image captured by a camera at a current time, determining that the image depicts a change in a region from a previous image captured by the camera at a previous time, determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of a known object type, determining, based on the determination that the change depicted in the image is of a known object type, whether the change does not correspond to a previously detected event, and determining, based on the determination that the change does not correspond to a previously detected event, whether the images captured by the camera between the current time and the previous time depict an event.

First claim

What is claimed is:

1. A computer-implemented method comprising:
maintaining, in computer memory, a background image that depicts a scene of a region, was captured by a camera at a first time, and is used to determine an action to perform given a determined change in the scene of the region;
obtaining an image depicting the region and captured by the camera at a second time that is different from the first time;
determining that the image depicts a change in the scene of the region from the background image, wherein determining that the image depicts the change comprises comparing the image to the background image;
determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of a known object type;
determining, based on the determination that the change depicted in the image is of a known object type, whether the change does not correspond to a previously detected event; and
in response to determining that the change does correspond to a previously detected event, generating an updated background image for the region.

2. The method of claim 1 , wherein determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of a known object type comprises determining that the change is of a package type.

3. The method of claim 1 , wherein obtaining the image depicting the region and captured by the camera at the second time comprises:
determining that a human is not detected in the image.

4. The method of claim 1 , comprising:
in response to determining that the change does correspond to a previously detected event, determining to use the updated background image to determine whether a future image depicts another change in the region.

5. A system comprising:
one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
maintaining, in computer memory, a background image that depicts a scene of a region, was captured by a camera at a first time, and is used to determine an action to perform given a determined change in the scene of the region;
obtaining an image depicting the region and captured by the camera at a second time that is different from the first time;
determining that the image depicts a change in the scene of the region from the background image, wherein determining that the image depicts the change comprises comparing the image to the background image;
determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of a known object type;
determining, based on the determination that the change depicted in the image is of a known object type and using first settings, whether the change does not correspond to a previously detected event; and
in response to determining that the change does not correspond to a previously detected event, repeating event detection with second settings, different from the first settings, during an interval between the first time and the second time.

6. The system of claim 5 , wherein determining, based on the determination that the change depicted in the image is of the known object type and using the first settings, whether the change does not correspond to the previously detected event comprises:
determining that an event was not detected between the first time and the second time; and
based on determining that an event was not detected between the first time and the second time, determining that the change does not correspond to a previously detected event.

7. The system of claim 6 , wherein determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of a known object type comprises determining that the change is of a package type, and
wherein determining that an event was not detected between the first time and the second time comprises determining whether a package delivery event was detected between the first time and the second time.

8. The system of claim 6 , wherein determining that an event was not detected between the first time and the second time comprises:
determining whether an event associated with a presence of a human was detected between the first time and the second time.

9. The system of claim 5 , wherein obtaining the image depicting the region and captured by the camera at the second time comprises:
determining that a human is not detected in the image.

10. The system of claim 5 , comprising:
based on determining that images captured by the camera between the second time and the first time depict an event, providing an indication of the event.

11. The system of claim 5 , wherein the operations comprise:
in response to repeating event detection with second settings, detecting an event of interest.

12. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
maintaining, in computer memory, a background image that depicts a scene of a region, was captured by a camera at a first time, and is used to determine an action to perform given a determined change in the scene of the region;
for each of a plurality of images:
obtaining the image depicting the region and captured by the camera at a second time that is different from the first time;
determining that the image depicts a change in the scene of the region from the background image, wherein determining that the image depicts the change comprises comparing the image to the background image;
determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of a known object type; and
determining, based on the determination that the change depicted in the image is of the known object type and using first settings, whether the change does not correspond to a previously detected event;

for a first image from the plurality of images and in response to determining that the change does correspond to a previously detected event, generating an updated background image for the region; and
for a second image from the plurality of images and in response to determining that the change does not correspond to a previously detected event, repeating event detection with second settings, different from the first settings, during an interval between the second time and the first time.

13. The computer-readable medium of claim 12 , wherein determining, based on the determination that the change depicted in the image is of the known object type and using the first settings, whether the change does not correspond to a previously detected event comprises:
determining whether an event was detected between the first time and the second time; and
based on determining that an event was not detected between the first time and the second time, determining that the change does not correspond to a previously detected event.

14. The computer-readable medium of claim 13 , wherein determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of the known object type comprises determining that the change is of a package type, and
wherein determining whether an event was detected between the first time and the second time comprises determining whether a package delivery event was detected between the first time and the second time.

Assignees

• Objectvideo Labs Llc

Inventors

• Ramanathan Narayanan
• Beach Allison
• Qian Gang
• Taheri Sima
• Yin Weihong
• Madden Donald Gerard

Classifications

• G06V20/52

  Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) · CPC title

• G06V40/10

  Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands · CPC title

• G06T7/248

  involving reference images or patches · CPC title

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title

• G06T2207/30232

  Surveillance · CPC title