Alert directives and focused alert directives in a behavioral recognition system
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.
1. A non-transitory, processor-readable medium storing instructions that, when executed by the processor, cause the processor to:
obtain characteristic values for an observed event in a scene, the characteristic values including at least one of: a pixel-height value, a pixel-width value, or an x-coordinate and y-coordinate center position of a foreground object,
parse a plurality of alert directives to identify a matching alert directive from the plurality of alert directives, each alert directive from the plurality of alert directives including at least one of a representation of whether a specified behavior should result in an alert, a representation of an alert publication frequency, or a representation specifying that matching to that alert directive is based on at least one of a behavior type or an object type, and
upon identifying the matching alert directive, bypass an existing alert publication procedure of the behavioral recognition system based on the matching alert directive.
2. The non-transitory, processor-readable medium of claim 1 , wherein the characteristic values include a set of x-coordinates and y-coordinates corresponding to a foreground object trajectory.
3. The non-transitory, processor-readable medium of claim 1 , wherein the matching alert directive includes a focus mask that intersects with a region in the scene where the observed event occurred.
4. The non-transitory, processor-readable medium of claim 1 , wherein the identification of the matching alert directive is based on at least one user-defined alert directive matching criterion.
5. The non-transitory, processor-readable medium of claim 1 , wherein at least one alert directive from the plurality of alert directives includes a pointer to an original alert on which that alert directive is based.
6. The non-transitory, processor-readable medium of claim 1 , wherein the one of publishing or not publishing an alert of the observed event in the scene is further based on a focus mask.
7. A method, comprising:
obtaining characteristic values for an observed event in a scene, the characteristic values including at least one of: a pixel-height value, a pixel-width value, or an x-coordinate and y-coordinate center position of a foreground object,
parsing a plurality of alert directives to identify a matching alert directive from the plurality of alert directives, each alert directive from the plurality of alert directives including at least one of a representation of whether a specified behavior should result in an alert, a representation of an alert publication frequency, or a representation specifying that matching to that alert directive is based on at least one of a behavior type or an object type, and
upon identifying the matching alert directive, bypassing an existing alert publication procedure of the behavioral recognition system based on the matching alert directive.
8. The method of claim 7 , wherein the characteristic values include a set of x-coordinates and y-coordinates corresponding to a foreground object trajectory.
9. The method of claim 7 , wherein the matching alert directive includes a focus mask that intersects with a region in the scene where the observed event occurred.
10. The method of claim 7 , wherein the identification of the matching alert directive is based on at least one user-defined alert directive matching criterion.
11. The method of claim 7 , wherein at least one alert directive from the plurality of alert directives includes a pointer to an original alert on which that alert directive is based.
12. The method of claim 7 , wherein the one of publishing or not publishing an alert of the observed event in the scene is further based on a focus mask.
13. A non-transitory, processor-readable medium storing instructions that, when executed by the processor, cause the processor to:
obtain characteristic values for an observed event in a scene,
parse a plurality of alert directives to identify a matching alert directive from the plurality of alert directives, each alert directive from the plurality of alert directives including (1) at least one of a representation of whether a specified behavior should result in an alert, a representation of an alert publication frequency, or a representation specifying that matching to that alert directive is based on at least one of a behavior type or an object type, (2) an identifier, (3) an alert pointer, and (4) an array of tolerance values for each alert characteristic from a plurality of alert characteristics, and
upon identifying the matching alert directive, bypass an existing alert publication procedure of the behavioral recognition system based on the matching alert directive.
14. The non-transitory, processor-readable medium of claim 13 , wherein the characteristic values include a set of x-coordinates and y-coordinates corresponding to a foreground object trajectory.
15. The non-transitory, processor-readable medium of claim 13 , wherein the matching alert directive includes a focus mask that intersects with a region in the scene where the observed event occurred.
16. The non-transitory, processor-readable medium of claim 13 , wherein the identification of the matching alert directive is based on at least one user-defined alert directive matching criterion.
17. The non-transitory, processor-readable medium of claim 16 , wherein the user-defined alert directive matching criterion is adjustable by a user via a graphical editor.
18. The non-transitory, processor-readable medium of claim 13 , wherein at least one alert directive from the plurality of alert directives includes a pointer to an original alert on which that alert directive is based.
19. The non-transitory, processor-readable medium of claim 13 , wherein the one of publishing or not publishing an alert of the observed event in the scene is further based on a focus mask.
20. The non-transitory, processor-readable medium of claim 19 , wherein the focus mask intersects with a region of the scene.