Body movement detector, body movement detection method and body movement detection program

• US 7,983,872 B2
• Filed: 03/09/2007
• Issued: 07/19/2011
• Est. Priority Date: 03/14/2006
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Makino, Kenichi, Takai, Motoyuki, Sano, Akane
• Primary Examiner(s): Dunn; Drew A
• Assistant Examiner(s): Cherry; Stephen J
• Application Number:
  US12/279,557
• Publication Number:
  US 20090240461A1
• Time in Patent Office: 1,593 Days
• Field of Search: 702/141
• US Class Current: 702/141
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/1123 Discriminating type of movement, e.g. walking or running A61B5/1116, A61B5/112 take precedence
• CPC Class Codes: A61B 5/7214 using signal cancellation, e.g. based on input of two identical physiological sensors spaced apart, or based on two signals derived from the same sensor, for different optical wavelengths
• CPC Class Codes: G01C 22/006 Pedometers

Abstract

A vertical component is extracted from an acceleration vector corresponding to body movement of a user from a three-axis acceleration sensor, and the vertical-component is separated by a high-band/low-band separating unit into a high-band component and a low-band component. Using these components, a peak detection/determination processing unit detects a peak-position candidate of action of the user in the vertical direction, identifies a peak-position candidate on the basis of an energy ratio between the high-band component and the low-band component, performs waveform matching regarding a predetermined range including each peak-position candidate, thereby determining a peak position, and detects body movement on the basis of the peak position, and a step-position analyzing unit detects a body-movement pitch.

Claims

1. A body-movement detecting apparatus comprising:

an acceleration sensor configured to be mounted on a body of a user;

vertical-component extracting means for extracting a vertical component of an acceleration from a detection output from the acceleration sensor;

separating means for performing component separation of the vertical component extracted by the vertical-component extracting means into a high-band component and a low-band component;

detecting means for detecting a peak-position candidate on the basis of the low-band component of the vertical component separated by the separating means;

identifying means for identifying the peak-position candidate detected by the detecting means as a peak-position candidate in a case where a ratio between energy of the low-band component and energy of the high-band component in a predetermined range including the peak-position candidate is less than a predetermined value; and

body-movement detecting means for detecting body movement of the user on the basis of the peak-position candidate identified by the identifying means, said body-movement detecting means counting a number of peak-position position candidates to determine a number of body movements of the user.

2. The body-movement detecting apparatus according to claim 1, further comprising:

deciding means for setting, for each peak-position candidate identified by the identifying means, a predetermined range including the peak-position candidate, performing matching with a waveform in another predetermined range using a waveform in each set predetermined range as a subject, and deciding the peak-position candidate included in the waveform used as the subject as a peak position in a case where a degree of similarity is high,

wherein the body-movement detecting means detects body movement of the user on the basis of the peak position decided by the deciding means.

3. The body-movement detecting apparatus according to claim 1, further comprising:

interval estimating means for estimating a step interval by performing analysis regarding a time-series pattern formed of a plurality of peak-position candidates identified by the identifying means or a time-series pattern formed of a plurality of peak positions decided by the deciding means; and

discriminating means for discriminating at least three states of “

stationary”

, “

walking/running”

, and “

undefined”

as action states of the user on the basis of the step interval estimated by the interval estimating means.

4. The body-movement detecting apparatus according to claim 3, wherein:

when the state is “

undefined”

, which is different from the “

stationary”

state, where no peak position is detected at all, and in which the step interval is not regular, determining means is capable of changing the action state of the user from “

undefined”

to “

walking/running”

in a case where the step interval is estimated by the interval estimating means as being regular during a number of steps, and

reference-pitch calculating means is provided for calculating a reference pitch on the basis of the step interval estimated by the interval estimating means in a case where the action state of the user has been changed by the determining means from “

undefined”

to “

walking/running”

.

5. The body-movement detecting apparatus according to claim 4, wherein:

the determining means compares a most recent step interval estimated by the interval estimating means with the reference pitch, and maintains the state of “

walking/running”

in a case where a difference thereof is within a preset error range or within a predetermined error range relative to an integer multiple of the reference pitch.

6. The body-movement detecting apparatus according to claim 4, further comprising:

counting means for counting steps of the user on the basis of the step interval estimated by the interval estimating means,

wherein the counting means performs interpolation of count on the basis of an integer multiple of the reference pitch of the step interval in a case where the action state of the user is discriminated as “

walking/running”

by the discriminating means.

7. The body-movement detecting apparatus according to claim 4, further comprising:

determining means for determining whether the action state of the user is “

walking”

or “

running”

on the basis of the step interval estimated by the interval estimating means in a case where the action state of the user is discriminated by the discriminating means as the state of “

walking/running”

.

8. The body-movement detecting apparatus according to claim 1, further comprising:

determining means for determining whether the action state of the user is “

walking”

or “

running”

on the basis of a peak value of the low-band component obtained by the obtaining means and the step interval estimated by the interval estimating means in a case where the action state of the user is discriminated by the discriminating means as the state of “

walking/running”

.

9. The body-movement detecting apparatus according to claim 3, further comprising:

piece-of-music playlist maintaining means for storing and maintaining one or more piece-of-music playlists formed in association with action states of the user; and

selecting means for selecting a piece-of-music playlist used to play a piece of music from the piece-of-music playlist maintaining means in accordance with an action state of the user discriminated by the discriminating means.

10. The body-movement detecting apparatus according to claim 1, further comprising:

playing means for piece-of-music data;

playing controlling means for controlling a playing rate of a piece of music by the playing means in accordance with a period of body movement of the user detected by the body-movement detecting means,

wherein the body-movement detecting apparatus has a function as a piece-of-music playing apparatus.

11. The body-movement detecting apparatus according to claim 1, wherein:

the acceleration sensor is of a multi-axis type, and

the vertical-component extracting means calculates a gravitational acceleration vector from an acceleration vector that is a detection output from the multi-axis acceleration sensor, and extracts a vertical component of an acceleration by performing calculation using the acceleration vector from the multi-axis acceleration sensor and the calculated gravitational acceleration vector.

12. A body-movement detecting method comprising:

a vertical-component extracting process of extracting a vertical component of an acceleration from a detection output of the acceleration sensor configured to be mounted on a body of a user;

a separating process of performing component separation of the vertical component extracted in the vertical-component extracting process into a high-band component and a low-band component;

a detecting process of detecting a peak-position candidate on the basis of the low-band component of the vertical component separated in the separating process;

an identifying process of identifying the peak-position candidate detected in the detecting process as a peak-position candidate in a case where a ratio between energy of the low-band component and energy of the high-band component in a predetermined range including the peak-position candidate is less than a predetermined value; and

a body-movement detecting process of detecting body movement of the user on the basis of the peak-position candidate identified in the identifying process, said body-movement detecting process counting a number of peak-position position candidates to determine a number of body movements of the user.

13. The body-movement detecting method according to claim 12, further comprising:

a deciding process of setting, for each peak-position candidate identified in the identifying process, a predetermined range including the peak-position candidate, performing matching with a waveform in another predetermined range using a waveform in each set predetermined range as a subject, and deciding the peak-position candidate included in the waveform used as the subject as a peak position in a case where a degree of similarity is high,

wherein, in the body-movement detecting process, body movement of the user is detected on the basis of the peak position decided in the deciding process.

14. The body-movement detecting method according to claim 12, further comprising:

an interval estimating process of estimating a process interval by performing analysis regarding a time-series pattern formed of a plurality of peak-position candidates identified in the identifying process or a time-series pattern formed of a plurality of peak positions decided in the deciding process; and

a discriminating process of discriminating at least three states of “

stationary”

, “

walking/running”

, and “

undefined”

as action states of the user on the basis of the process interval estimated in the interval estimating process.

15. The body-movement detecting method according to claim 12, wherein:

the acceleration sensor configured to be mounted on the body of the user is of a multi-axis type, and

in the vertical-component extracting process, a gravitational acceleration vector is calculated from an acceleration vector that is a detection output from the multi-axis acceleration sensor, and a vertical component of an acceleration is extracted by performing calculation using the acceleration vector from the multi-axis acceleration sensor and the calculated gravitational acceleration vector.

16. A non-transitory computer readable medium encoded with a body-movement detecting program causing a computer, the computer being mounted on a body-movement detecting apparatus that is provided with an acceleration sensor configured to be mounted on a body of a user and that detects body movement of the user by using a detection output from the acceleration sensor, to execute a method comprising:

a vertical-component extracting step of extracting a vertical component of an acceleration from the detection output of the acceleration sensor;

a separating step of performing component separation of the vertical component extracted in the vertical-component extracting step into a high-band component and a low-band component;

a detecting step of detecting a peak-position candidate on the basis of the low-band component of the vertical component separated in the separating step;

an identifying step of identifying the peak-position candidate detected in the detecting step as a peak-position candidate in a case where a ratio between energy of the low-band component and energy of the high-band component in a predetermined range including the peak-position candidate is less than a predetermined value; and

a body-movement detecting step of detecting body movement of the user on the basis of the peak-position candidate identified in the identifying step, said body-movement detecting step counting a number of peak-position position candidates to determine a number of body movements of the user.

17. The non-transitory computer readable medium according to claim 16, wherein the method further comprises:

a deciding step of setting, for each peak-position candidate identified in the identifying step, a predetermined range including the peak-position candidate, performing matching with a waveform in another predetermined range using a waveform in each set predetermined range as a subject, and deciding the peak-position candidate included in the waveform used as the subject as a peak position in a case where a degree of similarity is high,

wherein, in the body-movement detecting step, body movement of the user is detected on the basis of the peak position decided in the deciding step.

18. The non-transitory computer readable medium according to claim 16, wherein the method further comprises:

an interval estimating step of estimating a step interval by performing analysis regarding a time-series pattern formed of a plurality of peak-position candidates identified in the identifying step or a time-series pattern formed of a plurality of peak positions decided in the deciding step; and

a discriminating step of discriminating at least three states of “

stationary”

, “

walking/running”

, and “

undefined”

as action states of the user on the basis of the step interval estimated in the interval estimating step.

19. The non-transitory computer readable medium according to claim 16, wherein:

the acceleration sensor configured to be mounted on the body of the user is of a multi-axis type, and

in the vertical-component extracting step, a gravitational acceleration vector is calculated from an acceleration vector that is a detection output from the multi-axis acceleration sensor, and a vertical component of an acceleration is extracted by performing calculation using the acceleration vector from the multi-axis acceleration sensor and the calculated gravitational acceleration vector.

20. A body-movement detecting apparatus comprising:

an acceleration sensor configured to be mounted on a body of a user;

a vertical-component extractor configured to extract a vertical component of an acceleration from a detection output from the acceleration sensor;

a separating unit configured to separate components of the vertical component extracted by the vertical-component extractor into a high-band component and a low-band component;

a detector configured to detect a peak-position candidate on the basis of the low-band component of the vertical component separated by the separating unit;

an identifying unit configured to identify the peak-position candidate detected by the detector as a peak-position candidate in a case where a ratio between energy of the low-band component and energy of the high-band component in a predetermined range including the peak-position candidate is less than a predetermined value; and

a body movement detector configured to detect body movement of the user on the basis of the peak-position candidate identified by the identifying unit, said body-movement detector configured to count a number of peak-position position candidates to determine a number of body movements of the user.