VIDEO ANALYTICS METHOD AND SYSTEM

• US 20120194676A1
• Filed: 10/07/2010
• Published: 08/02/2012
• Est. Priority Date: 10/07/2009
• Status: Active Grant

• Current Assignee: iWatchLife Inc.
• Original Assignee: Telewatch, Inc.
• Inventors: Laganiere, Robert, Blais, Pascal, Phillips, Jason, Murphy, William
• Granted Patent: US 9,420,250 B2
• Time in Patent Office: Days
• US Class Current: 348/143
• CPC Class Codes: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44
• CPC Class Codes: G06V 20/44 Event detection
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: H04L 12/6418 Hybrid transport
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources
• CPC Class Codes: H04N 9/8042 involving data reduction

Abstract

A method for performing video analytics comprises capturing video data using a video source disposed at an acquisition end. The captured video data is stored at the acquisition end, and non-adjacent single frames of the captured video data are transmitted from the acquisition end to a server via an IP network. The non-adjacent single frames of captured video data are transmitted one at a time at known time intervals, or a burst of a plurality of non-adjacent single frames is transmitted at another known time interval. At the server, video analytics is performed on a current one of the non-adjacent single frames for detecting an actionable event relative to a previous one of the non-adjacent single frames. In response to detecting an actionable event, a signal for requesting previously stored video data captured between the current one of the non-adjacent single frames and the previous one of the non-adjacent single frames is transmitted to the acquisition end via the IP network.

Claims

1. A method comprising:

capturing video data using a video source disposed at an acquisition end;

storing the captured video data at the acquisition end;

transmitting non-adjacent single frames of the captured video data from the acquisition end to a server via an IP network, the non-adjacent single frames of captured video data being transmitted one at a time at known time intervals;

at the server, performing video analytics on a current one of the non-adjacent single frames for detecting an actionable event relative to a previous one of the non-adjacent single frames; and

,

in response to detecting an actionable event, transmitting to the acquisition end via the IP network a signal for requesting previously stored video data captured between the current one of the non-adjacent single frames and the previous one of the non-adjacent single frames.

2. A method according to claim 1, wherein the known time intervals between transmitting consecutive non-adjacent single frames are substantially longer than an interval of time between capturing two adjacent single frames of the video data.

4. A method according to claim 1, comprising transmitting from the acquisition end to a remote location via the IP network the requested previously stored video data for at least one of storage and processing thereof.

5. A method according to claim 4, comprising performing at the remote location video analytics on frames of the requested previously stored video data for determining a first frame containing video data relating to the detected actionable event.

6. A method according to claim 4, comprising displaying the requested previously stored video data for being viewed by a user.

7. A method according to claim 1, comprising providing an alarm signal in response to detecting the actionable event.

8. A method according to claim 1, comprising in response to detecting the actionable event, transmitting to the acquisition end via the IP network a signal for decreasing a duration of the time intervals between transmission of consecutive non-adjacent single frames.

9. A method according to claim 1, comprising in response to detecting the actionable event, transmitting to the acquisition end via the IP network a signal for requesting transmission of full frame rate video data from the video source disposed at the acquisition end.

3. (canceled)

10-11. -11. (canceled)

12. A method comprising:

capturing video data at a known frame rate using a video source disposed at au acquisition end;

transmitting non-adjacent single frames of the captured video data to a remote location via an IP network, the non-adjacent single frames of video data being transmitted at a first rate that is substantially slower than the known frame rate;

at a remote location, performing video analytics on each of the non-adjacent single frames of captured video data for detecting an actionable event; and

,

in response to detecting the actionable event, transmitting from a remote location to the acquisition end via the IP network a signal for increasing the rate of transmission of the non-adjacent single frames of video data to a second rate.

13. A method according to claim 12, comprising providing an alarm signal in response to detecting the actionable event.

14. A method according to claim 12, wherein the first rate of transmitting the non-adjacent single frames of the captured video data is at least thirty times slower than the known frame rate.

17. A method according to claim 12, wherein the second rate is substantially the same as the known frame rate of capturing the video data.

15-16. -16. (canceled)

18. A method comprising:

providing a first video source at a first acquisition end, the first video source relating to a first client;

providing a second video source at a second acquisition end, the second video source relating to a second client;

providing a server that is in communication with the first and second video sources via an IP network;

using the server, performing video analytics on video data provided separately to the server via the IP network from each one of the first video source and the second video source, the video analytics performed for detecting actionable events within the provided video data; and

,

providing in a selective manner an alarm signal to one of the first client and the second client in response to detecting an actionable event within the video data provided from one of the first video source and the second video source, respectively.

19. A method according to claim 18, wherein the first video data is captured at a known first frame rate using the first video source and the second video data is captured at a known second frame rate using the second video source, and wherein non-adjacent single frames of the captured first video data are transmitted to the server via the IP network at a first transmission rate that is substantially slower than the known first frame rate and non-adjacent single frames of the captured second video data are transmitted to the server via the IP network at a second transmission rate that is substantially slower than the known second frame rate.

20. A method according to claim 19, wherein the first transmission rate is different than the second transmission rate.

21. A method according to claim 20, wherein the first transmission rate is selected for detecting a first type of actionable event and the second transmission rate is selected for detecting a second type of actionable event.

22. A method according to claim 18, wherein providing the server comprises providing the server at a location that is remote from both the first video source and the second video source.

23. A method according to claim 18 wherein the first acquisition end and the second acquisition end are at a same location.

24. A method comprising:

providing a server farm comprising a plurality of servers;

providing a video source at an acquisition end, the video source in communication with the server farm via the IP network;

capturing video data at a known frame rate using the video source at the acquisition end;

transmitting non-adjacent single frames of the captured video data to the server farm via the IP network;

performing video analytics on the non-adjacent single frames of the captured video data using the plurality of servers of the server farm; and

,

detecting an actionable event based on a result of the video analytics performed by at least one server of the plurality of servers of the server farm.

25. A method according to claim 24, wherein the non-adjacent single frames are transmitted at a first rate that is substantially slower than the known frame rate.

26. A method according to claim 24, wherein different servers of the plurality of servers of the server farm perform video analytics on the non-adjacent single frames of the captured video data using different video analytics processes.

27. A method according to claim 24, wherein different servers of the plurality of servers of the server farm perform video analytics on the non-adjacent single frames of the captured video data in parallel using a same video analytics process.

28. A method according to claim 24, wherein each different server of the plurality of servers of the server farm perform video analytics on a different one of the non-adjacent single frames of the captured video data using a same video analytics process.

29. A method according to claim 25, comprising in response to detecting the actionable event, transmitting to the acquisition end via the IP network a signal for increasing the rate of transmission of the non-adjacent single frames of video data to a second rate.

30-35. -35. (canceled)