METHOD AND APPARATUS FOR FALL DETECTION

• US 20110231145A1
• Filed: 11/19/2009
• Published: 09/22/2011
• Est. Priority Date: 11/28/2008
• Status: Active Grant

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Chen, Ningjiang
• Granted Patent: US 8,756,032 B2
• Time in Patent Office: Days
• US Class Current: 702/141
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/1117 Fall detection
• CPC Class Codes: A61B 5/7239 using differentiation including higher order derivatives
• CPC Class Codes: A61B 5/7242 using integration
• CPC Class Codes: G08B 21/0446 worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait

Abstract

The invention relates to a method and apparatus for fall detection. According to the invention, the method for detecting the fall of an object comprises the steps of acquiring (501) a plurality of time-varying accelerations associated with a movement of said object; computing (502) a plurality of respective velocities and displacements in the direction of gravity, based on the plurality of time-varying accelerations; determining (503) lead time and return time of the movement, based on the plurality of accelerations, velocities and displacements, wherein the lead time equals a time interval from an original point of the movement to an extreme point of the movement, and the return time equals a time interval from the extreme point of the movement to a final point of the movement; and determining (504) whether the fall of said object occurs, based on whether a peak value of the velocities is larger than a predefined threshold value and whether the lead time is longer than the return time.

Claims

1. A method for detecting the fall of an object, said method comprising the steps of:

acquiring (501) a plurality of time-varying accelerations associated with a movement of said object;

computing (502) a plurality of respective velocities and displacements in the direction of gravity, based on the plurality of time-varying accelerations;

determining (503) lead time and return time of the movement, based on the plurality of accelerations, velocities and displacements, wherein the lead time equals a time interval from an original point (P₁) of the movement to an extreme point (E) of the movement, and the return time equals a time interval from the extreme point (E) of the movement to a final point (R) of the movement; and

determining (504) whether the fall of said object occurs, based on whether a peak value of the velocities is larger than a predefined threshold value and whether the lead time is longer than the return time.

2. A method as claimed in claim 1, wherein the extreme point (E) of the movement is a former point between a time point (V) corresponding to the peak value of the plurality of velocities in the direction of gravity and a time point (A) corresponding to a peak value of the acceleration amplitudes.

3. A method as claimed in claim 2, wherein the original point (P₁) of the movement is a time point preceding and being closest to the extreme point (E) of the movement where velocity equals zero.

4. A method as claimed in claim 3, wherein the final point (R) of the movement is a former point between a time point (P₂) succeeding and being closest to the extreme point (E) of the movement where velocity equals zero and a time point (D) corresponding to a peak value of the displacements in the direction of gravity.

5. A method as claimed in claim 1, further comprising the steps of:

computing (605) smoothness of time-varying velocities based on the plurality of velocities;

determining (606) whether the smoothness is larger than a predefined smoothness threshold; and

determining (607) that the fall of said object occurs when the smoothness is larger than the predefined smoothness threshold and the lead time is longer than the return time.

6. A method as claimed in claim 5, wherein the smoothness is given by the 2nd-order derivative of the plurality of velocities and is computed as:

7. A method as claimed in claim 6, wherein the accelerations include accelerations a_x, a_y, a_z, in x, y, z directions, respectively, and the velocities are given by

v=√

{square root over ((v_x)²+(v_y)²+(v_z)²)}{square root over ((v_x)²+(v_y)²+(v_z)²)}{square root over ((v_x)²+(v_y)²+(v_z)²)}−

∫

Gdt

wherein v_x=∫

a_xdt, v_y=∫

a_ydt, v_z=∫

a_zdt, and G is the acceleration of gravity.

8. An apparatus for detecting the fall of an object, said apparatus comprising:

an acquiring unit (801) configured to acquire a plurality of time-varying accelerations associated with a movement of said object;

a first computing unit (802) configured to compute a plurality of respective velocities and displacements in the direction of gravity, based on the plurality of time-varying accelerations;

a first determining unit (803) configured to determine lead time and return time of the movement, based on the plurality of accelerations, velocities and displacements, wherein the lead time equals a time interval from an original point (P₁) of the movement to an extreme point (E) of the movement, and the return time equals a time interval from the extreme point (E) of the movement to a final point (R) of the movement; and

a second determining unit (804) configured to determine whether a fall of said object occurs, based on whether a peak value of the velocities is larger than a predefined threshold value and whether the lead time is longer than the return time.

9. An apparatus as claimed in claim 8, wherein the extreme point (E) of the movement is a former point between a time point (V) corresponding to the peak value of the plurality of velocities in the direction of gravity and a time point (A) corresponding to a peak value of the acceleration amplitudes.

10. An apparatus as claimed in claim 9, wherein the original point (P₁) of the movement is a time point preceding and being closest to the extreme point (E) of the movement where velocity equals zero, and the final point (R) of the movement is a former point between a time point (P₂) succeeding and being closest to the extreme point (E) of the movement, where velocity equals zero, and a time point (D) corresponding to a peak value of the displacements in the direction of gravity.

11. An apparatus as claimed in claim 8, further comprising:

a second computing unit (905) configured to compute smoothness of time-varying velocities based on the plurality of velocities;

a third determining unit (906) configured to determine whether the smoothness is larger than a predefined smoothness threshold; and

wherein the second determining unit (804) is further configured to determine that the fall of said object occurs when the smoothness is larger than the predefined smoothness threshold and the lead time is longer than the return time.

12. An apparatus as claimed in claim 8, wherein the acquiring unit (801) is a 3D accelerometer.

13. An apparatus as claimed in claim 8, further comprising a preprocessor for preprocessing the plurality of accelerations, velocities and displacements.

14. A computer program product for detecting the fall of an object, comprising instructions being executable for performing the steps of:

acquiring (501) a plurality of time-varying accelerations associated with a movement of said object;

computing (502) a plurality of respective velocities and displacements in the direction of gravity, based on the plurality of time-varying accelerations;

determining (503) lead time and return time of the movement, based on the plurality of accelerations, velocities and displacements, wherein the lead time equals a time interval from an original point (P₁) of the movement to an extreme point (E) of the movement, and the return time equals a time interval from the extreme point (E) of the movement to a final point (R) of the movement; and

determining (504) whether the fall of said object occurs, based on whether a peak value of the velocities is larger than a predefined threshold value and whether the lead time is longer than the return time.

15. A computer program product as claimed in claim 14, wherein the extreme point (E) of the movement is a former point between a time point (V) corresponding to the peak value of the plurality of velocities in the direction of gravity and a time point (A) corresponding to a peak value of the acceleration amplitudes.