Alert directives and focused alert directives in a behavioral recognition system

• US 10,096,235 B2
• Filed: 03/15/2013
• Issued: 10/09/2018
• Est. Priority Date: 03/15/2012
• Status: Active Grant

• Current Assignee: Intellective Ai, Inc.
• Original Assignee: Omni AI, Inc.
• Inventors: Cobb, Wesley Kenneth, Seow, Ming-Jung, Xu, Gang, Saitwal, Kishor Adinath, Akins, Anthony, Joseph, Kerry, Urech, Dennis G.
• Primary Examiner(s): Atala, Jamie
• Assistant Examiner(s): Billah, Masum
• Application Number:
  US13/839,587
• Publication Number:
  US 20130242093A1
• Time in Patent Office: 2,034 Days
• Field of Search: 343143
• CPC Class Codes: G06F 18/41 Interactive pattern learning with a human teacher
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/0409 Adaptive resonance theory [ART] networks
• CPC Class Codes: G06N 3/088 Non-supervised learning, e.g. competitive learning
• CPC Class Codes: G06V 10/7788 the supervisor being a human, e.g. interactive learning with a human teacher
• 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: G06V 20/53 Recognition of crowd images, e.g. recognition of crowd congestion
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition
• CPC Class Codes: G08B 13/19608 Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
• CPC Class Codes: G08B 21/182 Level alarms, e.g. alarms responsive to variables exceeding a threshold
• CPC Class Codes: G08B 23/00 Alarms responsive to unspecified undesired or abnormal conditions
• CPC Class Codes: G08B 29/185 Signal analysis techniques for reducing or preventing false alarms or for enhancing the reliability of the system
• CPC Class Codes: H04N 7/002 Special television systems not provided for by H04N7/007 - H04N7/18 still pictures via a television channel H04N1/00098

Abstract

Alert directives and focused alert directives allow a user to provide feedback to a behavioral recognition system to always or never publish an alert for certain events. Such an approach bypasses the normal publication methods of the behavioral recognition system yet does not obstruct the system'"'"'s learning procedures.

Claims

1. A method for processing events generated from an evaluation of a stream of video frames, the method comprising:

obtaining characteristic values for an observed event in a scene depicted by the stream of video frames;

updating a learned state of the scene based on the characteristic values, wherein the learned state provides a model of patterns of behavior generated from evaluating a plurality of foreground objects detected in the stream of video frames;

parsing a list of alert directives for a matching alert directive having ranges of criteria values, wherein the characteristic values are within the ranges of the criteria values and wherein the alert directive overrides a decision to either publish an alert or to not publish an alert for the observed event based on the updated learned state of the scene without changing the updated learned state of the scene; and

upon identifying the matching alert directive, either publishing the alert or not publishing the alert according to alert directive.

2. The method of claim 1, wherein the observed event characteristic values are a pixel-height value, a pixel-width value, and an x- and y-coordinate center position of a foreground object.

3. The method of claim 1, wherein the observed event characteristic values are a set of x- and y-coordinates corresponding to a foreground object trajectory.

4. The method of claim 1, wherein the matching alert directive has a focus mask, the focus mask intersecting with a region in video frames of the scene where the observed event occurred.

5. The method of claim 1, wherein the criteria values may be adjusted in a graphical editor.

6. The method of claim 1, wherein publishing the alert regarding the observed event comprises generating one or more alerts based on the matching alert directive.

7. The method of claim 1, wherein parsing a list of alert directives for a matching alert directive comprises matching a first alert directive with an object type associated with a foreground object associated with the observed event and a behavior associated with the observed event.

8. A computer-readable storage medium storing instructions, which, when executed on a processor, performs an operation for processing events generated by a video analytics system evaluating a stream of video frames alerting a user to behavior corresponding to an alert directive, the operation comprising:

obtaining characteristic values for an observed event in a scene, depicted by the stream of video frames;

updating a learned state of the scene based on the characteristic values, wherein the learned state provides a model of patterns of behavior generated from evaluating a plurality of foreground objects detected in the stream of video frames;

parsing a list of alert directives for a matching alert directive having ranges of criteria values, wherein the characteristic values are within the ranges of the criteria values and wherein the alert directive overrides a decision to either publish an alert or to not publish an alert for the observed event based on the updated learned state of the scene without changing the updated learned state of the scene; and

upon identifying the matching alert directive, either publishing the alert or not publishing the alert regarding the observed event according to alert directive.

9. The computer-readable storage medium of claim 8, wherein the observed event characteristic values are a pixel-height value, a pixel-width value, and an x- and y-coordinate center position of a foreground object.

10. The computer-readable storage medium of claim 8, wherein the observed event characteristic values are a set of x- and y-coordinates corresponding to a foreground object trajectory.

11. The computer-readable storage medium of claim 8, wherein the matching alert directive has a focus mask, the focus mask intersecting with a region in video frames of the scene where the observed event occurred.

12. The computer-readable storage medium of claim 8, wherein the criteria values may be adjusted in a graphical editor.

13. The computer-readable storage medium of claim 8, wherein publishing the alert for the observed event comprises generating one or more alerts based on the matching alert directive.

14. The computer-readable storage medium of claim 8, wherein parsing a list of alert directives for a matching alert directive comprises matching a first alert directive with an object type associated with a foreground object associated with the observed event and a behavior associated with the observed event.

15. A system comprising:

a processor and

a memory hosting an application, which, when executed on the processor, performs an operation for processing events generated from an evaluation of a stream of video frame, the operation comprising;

obtaining characteristic values for an observed event in a scene depicted by the stream of video frames,

updating a learned state of the scene based on the characteristic values, wherein the learned state provides a model of patterns of behavior generated from evaluating a plurality of foreground objects detected in the stream of video frames,

parsing a list of alert directives for a matching alert directive having ranges of criteria values, wherein the characteristic values are within the ranges of the criteria values and wherein the alert directive overrides a decision to either publish an alert or to not publish an alert for the observed event based on the updated learned state of the scene without changing the updated learned state of the scene, and

upon identifying the matching alert directive, either publishing the alert or not publishing the alert regarding the observed event according to alert directive.

16. The system of claim 15, wherein the observed event characteristic values are a pixel-height value, a pixel-width value, and an x- and y-coordinate center position of a foreground object.

17. The system of claim 15, wherein the observed event characteristic values are a set of x- and y-coordinates corresponding to a foreground object trajectory.

18. The system of claim 15, wherein the matching alert directive has a focus mask, the focus mask intersecting with a region in video frames of the scene where the observed event occurred.

19. The system of claim 15, wherein the criteria values may be adjusted in a graphical editor.

20. The system of claim 15, wherein publishing the alert for the observed event comprises generating one or more alerts based on the matching alert directive.

21. The system of claim 15, wherein parsing a list of alert directives for a matching alert directive comprises matching a first alert directive with an object type associated with a foreground object associated with the observed event and a behavior associated with the observed event.