Apparatus and method for recognizing representative user behavior based on recognition of unit behaviors

Patent metadata

Field	Value
Publication number	US-10332011-B2
Application number	US-201514695806-A
Country	US
Kind code	B2
Filing date	Apr 24, 2015
Priority date	Jul 6, 2012
Publication date	Jun 25, 2019
Grant date	Jun 25, 2019

Abstract

An apparatus for recognizing a representative user behavior includes a unit-data extracting unit configured to extract at least one unit data from sensor data, a feature-information extracting unit configured to extract feature information from each of the at least one unit data, a unit-behavior recognizing unit configured to recognize a respective unit behavior for each of the at least one unit data based on the feature information, and a representative-behavior recognizing unit configured to recognize at least one representative behavior based on the respective unit behavior recognized for each of the at least one unit data.

First claim

What is claimed is:

1. An apparatus for recognizing a representative behavior of a user, the apparatus comprising:
one or more processors configured to:
generate plural unit data by performing a first dividing of sensor data, measured from plural sensors of different types, into first plural sections, and by performing a second dividing of the first plural sections into second plural sections,
where the performing of the second dividing further includes overlapping at least one section of the second plural sections with two consecutive sections of the first plural sections, and the generated plural unit data include the first plural sections and the second plural sections;
recognize unit behaviors of the user based on the generated plural unit data, the generated plural unit data including; information about an external environment of the user and information about the user; and
the representative behavior of the user based on a determined sequence of the recognized unit behaviors,
wherein the representative behavior comprises the sequence of the unit behaviors.

2. The apparatus of claim 1 , wherein the one or more processors are further configured to indicate a result of the recognition of the representative behavior, including controlling the apparatus based on the representative behavior being recognized as the user dozing off while driving.

3. The apparatus of claim 2 , wherein the one or more processors are further configured to indicate a result of the recognition of the representative behavior, including indicating an alert the user in response to the recognized representative behavior being recognized as the user dozing off while driving.

4. The apparatus of claim 3 , wherein the alert includes alerting the user by sending either one or both of a signal and a message to the user.

5. The apparatus of claim 1 , wherein the generated plural unit data each have a predetermined time length or a predetermined data size.

6. The apparatus of claim 1 , wherein the sensor data comprises sensor data measured from the plural sensors including any two or more of an acceleration sensor, a gyroscope, a Global Positioning System (GPS), a microphone, a light sensor, a camera, a proximity sensor, a touch panel, a software sensor, and a biosensor.

7. The apparatus of claim 1 ,
wherein the one or more processors are further configured to extract feature information from the generated plural unit data, and
for the recognizing of the unit behaviors, one or more processors are configured to recognize the unit behaviors based on the extracted feature information.

8. The apparatus of claim 7 ,
wherein, for generating of the plural unit data, the one or more processors are configured to read time-domain unit data from the sensor data, and convert the time-domain unit data into frequency-domain unit, and
for the extracting of the feature information, the one or more processors are configured to extract the feature information from the frequency-domain unit data.

9. The apparatus of claim 7 , wherein the feature information comprises any one or any combination of any two or more of a mean, entropy, energy, and a correlation.

10. The apparatus of claim 7 , wherein, for generating of the plural unit data, one or more processors are further configured to classify the feature information into the unit behaviors recognized from the generated plural unit data, using a predetermined learning algorithm learned from learning data.

11. The apparatus of claim 10 , wherein the predetermined learning algorithm is any one of a Bayesian learning algorithm, a Support Vector Machine (SVM) learning algorithm, and a Decision Tree learning algorithm.

12. The apparatus of claim 10 , wherein the one or more processors are further configured to update the learning data based on the unit behaviors recognized from the generated plural unit data, or the representative behavior, or the unit behaviors recognized from the generated plural unit data and the representative behavior.

13. The apparatus of claim 1 , wherein, for the recognizing of the representative behavior, the one or more processors are configured to classify the unit behaviors recognized from the generated plural unit data into the representative behavior, using any one of a Bayesian learning algorithm, a Support Vector Machine (SVM) learning algorithm, a Decision Tree learning algorithm, a Hidden Markov Model (HMM) learning algorithm, and a Finite-State Machine (FSM) learning algorithm.

14. The apparatus of claim 1 , wherein, for the recognizing of the representative behavior, the one or more processors are configured to select at least one unit behavior from the unit behaviors recognized from the generated plural unit data based on a predetermined criterion, and classify a representative behavior of the selected at least one unit behavior as a representative user behavior for the unit behaviors.

15. The apparatus of claim 1 , wherein each of the unit behaviors is a recognized movement of the user or a recognized manner of speech of the user.

16. The apparatus of claim 1 , wherein the one or more processors are further configured to indicate a result of the recognition of the representative behavior.

17. A processor-implemented method, performed by a computing apparatus using one or more processors, of recognizing a representative behavior of a user, the method comprising:
generating, using the one or more processors of the computing apparatus, plural unit data by performing a first dividing of sensor data, measured from plural sensors of different types, into first plural sections, and by performing a second dividing of the first plural sections into second plural sections, where
the performing of the second dividing further includes overlapping at least one section of the second plural sections with two consecutive sections of the first plural sections, and the generated plural unit data include the first plural sections and the second plural sections;
recognizing, using the one or more processors, unit behaviors of the user based on the generated plural unit data, the generated plural unit data including information about an external environment of the user and information about the user; and
recognizing, using the one or more processors, representative behavior of the user based on a determined sequence of the recognized unit behaviors,
wherein the representative behavior comprises the sequence of the unit behaviors.

18. The method of claim 17 , further comprising indicating a result of the recognition of the representative behavior, including controlling the apparatus based on the recognized representative behavior being recognized as the user dozing off while driving.

19. The method of claim 18 , further comprising indicating a result of the recognition of the representative behavior, including indicating an alerting the user in response to the recognized representative behavior being recognizing as the user dozing off while driving.

20. The method of claim 19 , wherein the alerting of the user comprises alerting the user by sending either one or both of a signal and a message to the user.

21. The method of claim 17 , wherein the generated plural unit data each have a predetermined time length or a predetermined data size.

22. The method of claim 17 , wherein the sensor data comprises sensor data measured from the plural sensors including any two or more of an acceleration sensor, a gyroscope, a Global Positioning System (GPS), a microphone, a l

Assignees

• Samsung Electronics Co Ltd

Inventors

• Kim Hyun-Jun

Classifications

• G06N5/025 Primary

  Extracting rules from data · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06N5/00

  Computing arrangements using knowledge-based models · CPC title

• G06F17/00 Primary

  Digital computing or data processing equipment or methods, specially adapted for specific functions (information retrieval, database structures or file system structures therefor G06F16/00) · CPC title