Method and apparatus for detecting an event based on patterns of behavior
A system and apparatus are disclosed for modeling patterns of behavior of humans or other animate objects and detecting a violation of a repetitive pattern of behavior. The behavior of one or more persons is observed over time and features of the behavior are recorded in a multi-dimensional space. Over time, the multi-dimensional data provides an indication of patterns of human behavior. Activities that are repetitive in terms of time, location and activity, such as sleeping and eating, would appear as a Gaussian distribution or cluster in the multi-dimensional data. Probability distribution functions can be analyzed using known Gaussian or clustering techniques to identify repetitive patterns of behavior and characteristics thereof, such as a mean and variance. Deviations from repetitive patterns of behavior can be detected and an alarm can be triggered, if appropriate.
1. A method for detecting an event based on behavior of a person, said method comprising:
visually observing a plurality of behaviors of the person in a five dimensional space that includes vertical distance, horizontal distance, time, body posture and magnitude of body motion;
obtaining a plurality of images;
identifying at least one pattern of behavior from the visual observation of the plurality of behaviors;
recognizing the at least one pattern of behavior in said plurality of images by fitting a probability distribution function to said plurality of images; and
detecting an absence of said pattern of behavior.
2. The method of claim 1 , wherein said detecting further comprises generating an alarm.
3. The method of claim 1 , wherein said fitting identifies at least one Gaussian having a mean and variance.
4. The method of claim 3 , wherein said detecting further comprises determining if features extracted from said plurality of images exceed a predefined threshold distance based upon said variance of a Gaussian.
5. The method of claim 1 , wherein said recognizing further comprises clustering said plurality of images to identify said at least one repetitive pattern of behavior.
6. The method of claim 5 , wherein said clustering step further comprises computing a mean and variance of said clustered images.
7. The method of claim 6 , wherein said detecting further comprises determining if features extracted from said plurality of images exceed a predefined threshold distance based upon said variance of a cluster.
8. The method of claim 6 , wherein said detecting further comprises evaluating a distance between said features extracted from said plurality of images and one of said clustered images.
9. The method of claim 1 , further comprising extracting a plurality of features from said plurality of images.
10. The method of claim 9 , further comprising expressing said plurality of extracted features as a probability distribution function.
11. The method of claim 9 , further comprising expressing said plurality of extracted features in said five dimensional space.
12. A system for detecting an event based on behavior of a person, said system comprising:
a memory that stores computer-readable code; and
a processor operatively coupled to said memory, said processor configured to implement said computer-readable code, said computer-readable code configured to:
visually observe a plurality of behaviors of the person in a five dimensional space that includes vertical distance, horizontal distance, time, body posture and magnitude of body motion;
obtaining a plurality of images;
identify at least one pattern of behavior from the visual observation of the plurality of behaviors;
recognize the at least one pattern of behavior in said plurality of images by fitting a probability distribution function to said plurality of images; and
detect an absence of said pattern of behavior.
13. The system of claim 12 , wherein said processor is further configured to generate an alarm.
14. The system of claim 12 , wherein said processor is further configured to identify at least one Gaussian having a mean and variance.
15. The system of claim 14 , wherein said processor is further configured to determine if features extracted from said plurality of images exceed a predefined threshold distance based upon said variance of a Gaussian.
16. The system of claim 12 , wherein said processor is further configured to cluster said plurality of images to identify said at least one repetitive pattern of behavior.
17. The system of claim 16 , wherein said processor is further configured to compute a mean and variance of said clustered images.
18. The system of claim 17 , wherein said processor is further configured to determine if features extracted from said plurality of images exceed a predefined threshold distance based upon said variance of a cluster.
19. The system of claim 17 , wherein said processor is further configured to evaluate a distance between said features extracted from said plurality of images and one of said clustered images.
20. The system of claim 12 , wherein said processor is further configured to extract a plurality of features from said plurality of images.
21. The system of claim 20 , wherein said processor is further configured to express said plurality of extracted features as a probability distribution function.
22. An article of manufacture for detecting an event based on behavior of a person, said article of manufacture comprising:
a computer-readable medium having stored thereon computer-readable code to:
visually observe a plurality of behaviors of the person in a five dimensional space that includes vertical distance, horizontal distance, time, body posture and magnitude of body motion;
obtaining a plurality of images;
identify at least one pattern of behavior from the visual observation of the plurality of behaviors;
recognize the at least one pattern of behavior in said plurality of images by fitting a probability distribution function to said plurality of images to identify at least one Gaussian having a mean and variance; and
detect an absence of said pattern of behavior.