Systems and methods for video analysis rules based on map data

Patent metadata

Field	Value
Publication number	US-10708548-B2
Application number	US-201816139446-A
Country	US
Kind code	B2
Filing date	Sep 24, 2018
Priority date	Dec 5, 2014
Publication date	Jul 7, 2020
Grant date	Jul 7, 2020

Abstract

Systems, methods and computer-readable media for creating and using video analysis rules that are based on map data are disclosed. A sensor(s), such as a video camera, can track and monitor a geographic location, such as a road, pipeline, or other location or installation. A video analytics engine can receive video streams from the sensor, and identify a location of the imaged view in a geo-registered map space, such as a latitude-longitude defined map space. A user can operate a graphical user interface to draw, enter, select, and/or otherwise input on a map a set of rules for detection of events in the monitored scene, such as tripwires and areas of interest. When tripwires, areas of interest, and/or other features are approached or crossed, the engine can perform responsive actions, such as generating an alert and sending it to a user.

First claim

What is claimed is:

1. A computer implemented method to generate video analysis rules, comprising:
displaying, on a map based user interface, a map of a geographic area, at least a portion of the geographic area covered by a video sensor;
accepting, via the map based user interface, a rule-representing feature that is placed in a specific location on the map by a user;
generating a video analysis rule based on the rule representing feature and the specific location on the map, wherein the video analysis rule is expressed in a geo-registered map space;
receiving video from the video sensor; and
applying the video analysis rule to the video from the video sensor to detect an event.

2. The method of claim 1 , further comprising generating a notification on detection of the event.

3. The method of claim 1 , wherein the rule-representing feature is a tripwire.

4. The method of claim 1 , wherein the rule-representing feature is an area of interest.

5. The method of claim 1 , wherein the geo-registered map space is defined in coordinates that are latitude and longitude coordinates.

6. The method of claim 1 , wherein the video provided by the sensor comprises pixel-space video.

7. The method of claim 6 , further comprising:
transforming the pixel-space video into video in the geo-registered map space.

8. The method of claim 7 , wherein transforming further comprises:
associating an element from a frame of the video with coordinates in the geo-registered map space.

9. The method of claim 8 , wherein the coordinates in the geo-registered map space are latitude and longitude coordinates.

10. The method of claim 8 , wherein the video analysis rule is expressed in coordinates in the geo-registered map space; and
wherein applying the video analysis rule to the video from the sensor comprises:
comparing the coordinates of the element from the frame of the video with the coordinates of the video analysis rule to detect the triggering event.

11. A method for generating video analysis rules, the method comprising:
displaying, by a computing device, a map of a geographic area covered by a sensor that provides video of at least a portion of the geographic area;
providing, by the computing device, a user with an interface to place a rule-representing feature in a specific location on the map;
generating, by the computing device, a video analysis rule based on the rule-representing feature and the specific location on the map, wherein the video analysis rule is expressed in a geo-registered map space;
applying, by the computing device, the video analysis rule to the video from the sensor to detect an event in the video; and
generating, by the computing device, a notification upon detection of the event.

12. The method of claim 11 , wherein the notification causes the connected device to perform an action from the group consisting of: emitting a sound; displaying a view of the geographic area that is covered by the sensor; and providing a user-selectable link that activates transfer of information regarding the triggering event from another device.

13. The method of claim 11 , wherein the rule-representing feature is a tripwire.

14. The method of claim 11 , wherein the rule-representing feature is an area of interest.

15. The method of claim 11 , wherein the geo-registered map space is defined in coordinates that are latitude and longitude coordinates.

16. The method of claim 11 , wherein the video comprises pixel-space video.

17. The method of claim 16 , further comprising:
transforming the pixel-space video into video in the geo-registered map space.

18. The method of claim 17 , wherein transforming further comprises:
associating an element from a frame of the video with coordinates in the geo-registered map space.

19. The method of claim 18 , wherein the coordinates in the geo-registered map space are latitude and longitude coordinates.

20. The method of claim 18 , wherein the video analysis rule is expressed in coordinates in the geo-registered map space; and
wherein applying the video analysis rule to the video from the sensor comprises:
comparing the coordinates of the element from the frame of the video with the coordinates of the video analysis rule to detect the triggering event.

Assignees

• Avigilon Fortress Corp

Inventors

• Rasheed Zeeshan
• Eubanks Dana
• Yin Weihong
• Zhang Zhong
• Glowacki Kyle
• Beach Allison

Classifications

• G06V10/772

  Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries · CPC title

• G06F18/28

  Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title

• G06V20/52

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

• H04N7/181 Primary

  for receiving images from a plurality of remote sources · CPC title

• G08B13/19671

  Addition of non-video data, i.e. metadata, to video stream · CPC title