Visual rules for decision management

• US 9,201,581 B2
• Filed: 07/31/2013
• Issued: 12/01/2015
• Est. Priority Date: 07/31/2013
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adi, Asaf, Arcushin, Sarit, Bak, Peter, Dotan, Dolev, Kofman, Alexander, Yaeli, Avi, Yatzkar-Haham, Tali, Saadoun, Gilad, Sharon, Guy
• Primary Examiner(s): Bella, Matthew
• Assistant Examiner(s): SHIN, SOO JUNG
• Application Number:
  US13/955,005
• Publication Number:
  US 20150040064A1
• Time in Patent Office: 853 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 3/04817 using icons graphical or visual programming using iconic symbols G06F8/34
• CPC Class Codes: G06F 3/04842 Selection of displayed objects or displayed text elements G06F3/0482 takes precedence
• CPC Class Codes: G06F 3/04847 Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 50/06 Energy or water supply

Abstract

A computerized method for generating monitoring rules, comprising: presenting to a user a spatio-temporal data indicative of a plurality of spatio-temporal attributes of a plurality of image objects in a space during a period; selecting, by a user, at least one of a spatial pattern and a temporal pattern represented in said spatio-temporal data; analyzing said at least one of a spatial pattern and a temporal pattern to identify at least one of a spatial related characteristic of at least some of said plurality of image objects and a temporal related characteristic of at least some of said plurality of image objects; automatically generating at least one monitoring rule to track an occurrence of said at least one of a location related characteristic and a temporal related characteristic.

Claims

1. A computerized method for generating monitoring rules from a user marking of visual data, comprising:

presenting on a display a spatio-temporal data comprising images representing a plurality of physical objects in a space during a period;

while said spatio-temporal data is presented on said display, receiving a plurality of visual selections on said display from a user, said plurality of visual selections marking a region on said display;

analyzing said region in said images to identify a spatio-temporal pattern defining at least one relationship between some of said plurality of physical objects in said region, wherein said spatio-temporal pattern comprises at least one a location related characteristic and at least one temporal related characteristic of some of said plurality of physical objects in said region; and

automatically generating at least one quantifiable monitoring rule to track a following occurrence of said spatio-temporal pattern by identifying said at least one location related characteristic and said at least one temporal related characteristic of a new physical object represented in a new spatio-temporal data of said space in a following period.

2. The method of claim 1, wherein said spatio-temporal data is provided by a sequence of video images depicting movement of said plurality of physical objects in said space.

3. The method of claim 1, wherein said location related characteristic is a combination of a portion of a plurality of spatio-temporal attributes of said plurality of physical objects.

4. The method of claim 1, wherein said plurality of visual selections are determined by analyzing how a user visually explores said spatio-temporal data.

5. The method of claim 1, wherein said plurality of visual selections are performed by applying a member of a group consisting of:

a filtering condition on said spatio-temporal data, a filtering condition to said plurality of physical objects, a filtering condition to a plurality of spatio-temporal attributes of said plurality of physical objects, mapping at least one of said plurality of spatio-temporal attributes to said plurality of physical objects, coloring code mapped said at least one of said plurality of spatio-temporal attributes, mapping at least one of said plurality of spatio-temporal attributes to a visually distinct shape and generating a spatio-temporal attribute for said plurality of physical objects by relating at least a pair of said plurality of spatio-temporal attributes;

said plurality of visual selections is performed by a user performing at least one of;

selecting a sub area of said space during a certain time frame of said period, hovering over an object of said plurality of physical objects, clicking on an object of said plurality of physical objects and tagging an object of said plurality of physical objects; and

said analyzing is performed according to said plurality of visual selection.

6. The method of claim 1, wherein said plurality of visual selections is performed by a plurality of modifying actions and said automatically generating a plurality of monitoring rule elements is performed according to each action.

7. The method of claim 6, further comprising:

recognizing a plurality of relationships between said plurality of modifying actions;

filtering a sub-set of said plurality of modifying actions according to recognized said plurality of relationships.

8. The method of claim 1, further comprising:

applying said at least one monitoring rule to a new inspected set of video images.

9. The method of claim 1, further comprising:

applying said at least one monitoring rule to a plurality of dynamic physical objects, each having a plurality of changing spatio-temporal attributes.

10. The method of claim 1, further comprising:

defining, by a user, at least one condition for applying said at least one monitoring rule.

11. The method of claim 1, wherein said presenting on a display is triggered by availability of said spatio-temporal data.

12. The method of claim 1, wherein said automatically generating a plurality of monitoring rule elements is performed by extracting filter elements from a filter used for said selecting.

13. The method of claim 1, wherein said automatically generating a plurality of monitoring rule elements is performed by extracting at least one mapping attribute from a visual mapping used for said selecting.

14. The method of claim 1, wherein said automatically generating a plurality of monitoring rule elements is performed by extracting a second plurality of spatio-temporal attributes from said region of said image.

15. A computerized method for generating monitoring rules from a user marking of visual data, comprising:

a non-transitory computer readable storage medium;

first program instructions to present on a display, to a user, spatio-temporal data comprising images representing a plurality of spatio-temporal attributes of a plurality of physical objects in a space during a period;

second program instructions to enable a user to input a plurality of visual selections marking on said display a region;

third program instructions to analyze said region in said images to identify a spatio-temporal pattern defining at least one relationship between some of said plurality of physical objects in said region, wherein said spatio-temporal pattern comprises at least one a location related characteristic and at least one temporal related characteristic of some of said plurality of physical objects in said region

fourth program instructions to automatically generate a plurality of monitoring rule elements according to said plurality of visual selections;

fifth program instructions to automatically apply at least one monitoring rule created using at least one of said plurality of monitoring rule elements for tracking following occurrence of said spatio-temporal pattern by identifying said at least one of a location related characteristic and said at least one temporal related characteristic of a new physical object represented in a new spatio-temporal data of said space in a following period;

wherein said first, second, third, fourth, and fifth program instructions are stored on said non-transitory computer readable storage medium.

16. A system for generating monitoring rules from a user marking of visual data comprising:

a processor;

a user interface enabling a user to input a plurality of visual selections marking a region on a display identifying a spatio-temporal pattern represented in a spatio-temporal data while said spatio-temporal data is presented on said display;

a displaying unit which displays to a user said spatio-temporal data comprising images representing a plurality of physical objects in a space during a period; and

a rule element suggestion engine configured to instruct said processor to analyze said region in said images to identify said spatio-temporal pattern defining at least one of a location related characteristic and at least one temporal related characteristic of some of said plurality of physical objects in said region, and configured to instruct said processor to automatically generate a monitoring rule to track a following occurrence of said spatio-temporal pattern by identifying said at least one location related characteristic and said at least one temporal related characteristic of a new physical object represented in a new spatio-temporal image data of said space in a following period.

17. The system of claim 16, wherein said rule element suggestion engine is configured to instruct said processor to generate automatically a plurality of monitoring rule elements according to modified said selected said spatial-temporal pattern wherein said monitoring rule comprises at least one of said monitoring rule elements.

18. The system of claim 16, wherein said user interface enables a user to modify displayed said image and to create at least one monitoring rule using at least one of a plurality of monitoring rule elements.