Method and device for fall prevention and detection

• US 20060145874A1
• Filed: 11/21/2003
• Published: 07/06/2006
• Est. Priority Date: 11/21/2002
• Status: Active Grant

• Current Assignee: Secumanagement B.V.
• Original Assignee: Secumanagement B.V.
• Inventors: Fredriksson, Anders, Rosqvist, Fredrik
• Granted Patent: US 7,541,934 B2
• Time in Patent Office: Days
• US Class Current: 340/573.100
• CPC Class Codes: G08B 21/043 detecting an emergency event, e.g. a fall
• CPC Class Codes: G08B 21/0446 worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait
• CPC Class Codes: G08B 21/0461 integrated or attached to an item closely associated with the person but not worn by the person, e.g. chair, walking stick, bed sensor
• CPC Class Codes: G08B 21/0476 Cameras to detect unsafe condition, e.g. video cameras

Abstract

Method and device for fall prevention and detection, specially for the elderly care based on digital image analysis using an intelligent optical sensor. The fall detection is divided into two main steps; finding the person on the floor, and examining the way in which the person ended up on the floor. The first step is further divided into algorithms investigating the percentage share of the body on the floor, the inclination of the body and the apparent length of the person. The second step includes algorithms examining the velocity and acceleration of the person. When the first step indicates that the person is on the floor, data for a time period of a few seconds before and after the indication is analysed in the second step. If this indicates fall, a countdown state is initiated in order to reduce the risk of false alarms, before sending an alarm. The fall prevention is also divided into two main steps; identifying a person entering a bed, and identifying the person leaving the bed to end up standing beside it. The second step is again further divided into algorithms investigating the surface area of on or more objects in an image, the inclination and the apparent length of these objects. When the second step indicates that a person is in an upright condition, a countdown state is initiated in order to allow for the person to return to the bed.

Claims

1. A method of monitoring an object with respect to a potential fall condition, comprising:

observing a detection area with an optical detector;

determining, based on at least one image of the detection area, that an object is in an upright condition in the detection area;

waiting for a predetermined time period; and

emitting an alarm after said predetermined time period.

2. The method of claim 1, further comprising determining an angle between the object and a vertical direction, wherein the step of determining that the object is in an upright condition comprises determining that the angle is below 20 degrees, preferably below 10 degrees.

3. The method of claim 2, wherein said step of determining an angle comprises:

transforming foot image coordinates (u_f, v_f) of a foot portion of the object into foot room coordinates (X_f, Y_f=0, Z_f);

adding a length Δ

Y to a vertical coordinate of the foot room coordinates;

transforming at least the vertical coordinate to form top image coordinates (u_h,V_h), whereby the vertical direction is given by a vector between the foot image coordinates (u_f, V_f) and the top image coordinates (u_h,v_h); and

determining an angle between said vector and the object.

4. The method of claim 3, further comprising:

determining a direction of the object by calculating mass centres of at least two extreme parts of the object and determining a vector between them as the direction of the object.

5. The method of claim 1, further comprising:

determining mass centres of at least two extreme parts of the object; and

determining a length of the object;

wherein said step of determining that the object is in an upright condition comprises determining that the length of the object is above a predetermined length.

6. The method of claim 5, wherein said predetermined length represents an object length of at least 2 meters.

7. The method of claim 5, further comprising:

transforming said mass centres into room coordinates at a floor level (Y=0) in the detection area; and

determining the length of the object in said room coordinates.

8. The method of claim 1, further comprising:

defining a room height limit in room coordinates;

transforming said room height limit into an image height limit in image coordinates;

forming a foreground image by calculating a difference between a current image and a background image;

deriving a number of foreground elements from the foreground image, said number representing the foreground elements that are located below the image height limit in the foreground image;

wherein said step of determining that the object is in an upright condition comprises determining that said number exceeds a predetermined value.

9. The method of claim 1, further comprising calculating the surface area of the object;

wherein said step of determining that the object is in an upright condition comprises determining that the surface area exceeds a predetermined minimum value.

10. The method of claim 1, wherein a state for checking for an upright condition is initiated upon the identification of a movement in at least part of a bed in the detection area.

11. A device for monitoring an object with regard to a potential fall condition, comprising:

a detector for observing a detection area;

a determination device for determining, based on at least one image from the detector, that an object is in an upright condition in the detection area; and

an alarm device for emitting an alarm a predetermined time period after a determination of an upright condition by the determination device.

12. The device of claim 11, wherein said determination device further comprises an angle calculation device for calculating an angle between the object and a vertical direction, wherein determining that the object is in an upright condition comprises determining that the angle is below 20 degrees, preferably below 10 degrees.

13. The device of claim 11, wherein said determination device further comprises a length calculation device for determining mass centres of at least two extreme parts of the object;

and calculating a length of the object;

wherein determining that the object is in an upright condition comprises determining that the length of the object is above a predetermined length.

14. The device of claim 11, wherein said determination device further comprises a height limit calculation device for defining a room height limit in room coordinates;

transforming said room height limit into an image height limit in image coordinates;

forming a foreground image by calculating a difference between a current image and a background image;

deriving a number of foreground elements from the foreground image, said number representing the foreground elements that are located below the image height limit in the foreground image;

wherein determining that the object is in an upright condition comprises determining that said number exceeds a predetermined value.

15. The device of claim 11, wherein said determination device further comprises an area calculation device for calculating the surface area of the object;

wherein determining that the object is in an upright condition comprises determining that the surface area exceeds a predetermined minimum value.

16. The device of claim 11, further comprising of a movement detector for identifying movement in at least part of a bed in the detection area, wherein said determination device is initiated to check for an upright condition upon the identification of a movement by the movement detector.

20. The method of claim 16, further comprising:

determining an angle between the object and a vertical direction, wherein the step of determining that the object is lying on a floor comprises determining that the angle is above 10 degrees, preferably above 20 degrees.

21. The method of claim 20, wherein said step of determining an angle comprises:

transforming foot image coordinates (u_f, v_f) of a foot portion of the object into foot room coordinates (X_f, Y_f=0, Z_f);

adding a length Δ

Y to a vertical coordinate of the foot room coordinates;

transforming at least the vertical coordinate to form top image coordinates (u_h,v_h), whereby the vertical direction is given by a vector between the foot image coordinates (u_f, v_f) and the top image coordinates (u_h,v_h); and

determining an angle between said vector and the object.

22. The method of claim 21, further comprising:

determining a direction of the object by calculating mass centres of at least two extreme parts of the object and determining a vector between them as the direction of the object.

23. The method of claim 16, further comprising:

determining mass centres of at least two extreme parts of the object;

determining a length of the object;

wherein said step of determining that the object is lying on a floor comprises determining that the length of the object is below a predetermined length.

24. The method of claim 23, wherein said predetermined length represents an object length of less than 4 meters, such as below 3 meters, or specifically below 2 meters.

25. The method of claim 23, further comprising:

transforming said mass centres into room coordinates at a floor level (Y=0) in the detection area; and

determining the length of the object in said room coordinates.

26. The method of claim 16, further comprising:

deriving an image sequence for at least a time period preceding the determination that the object is lying on the floor; and

analysing the derived image sequence for high velocities and/or negative accelerations;

wherein a subsequent step for identifying a fall condition comprises determining that the velocity is above a predetermined value and/or the acceleration is below a negative value.

27. The method of claim 26, wherein said time period includes a time before and a time after said determination.

28. The method of claim 26, wherein said time period is 2 seconds.

29. The method of any claim 16, further comprising:

forming a foreground image by calculating a difference between a current image and a previous image; and

deriving a number of foreground elements from the foreground image;

wherein a subsequent step for identifying a fall condition comprises determining that the number of foreground elements exceeds a foreground number value.

30. The method of claim 29, wherein the foreground number value represents the number of foreground elements in a reference foreground image which is derived by calculating a difference between a current image and a background image.

31. The method of claim 29, further comprising:

defining a room height limit in room coordinates;

transforming said room height limit into an image height limit in image coordinates;

wherein said number of foreground elements represents the foreground elements that are located below the image height limit in the foreground image.

32. The method of claim 29, wherein said current image is set as said background image if there is no change in the foreground image during a predetermined time period.

33. The method of claim 26, further comprising:

pre-calculating a probability curve for a fall condition and a probability curve for a non-fall condition for velocity and/or negative acceleration, wherein a subsequent step for identifying a fall condition comprises determining that the velocity and/or the acceleration has the highest probability for a fall condition.

34. The method of claim 26, wherein the identification of the fall condition is initiated upon the determination that the object is lying on the floor.

17. A method of monitoring an object with regard to a fall condition, comprising:

observing a detection area with an optical detector;

determining, based on at least one image of the detection area, that an object is lying on a floor in the detection area;

waiting for a predetermined time period; and

emitting an alarm after said predetermined time period.

18. The method of claim 17, wherein said time period is more than 2 minutes, such as between 5 and 15 minutes, and more specifically about 10 minutes.

19. The method of claim 17, further comprising:

calculating a foreground image, which is the difference between a current image and a predetermined background image; and

calculating the ratio of the foreground image that is present on the floor of the detection area and the total foreground image;

wherein said step of determining that the object is lying on the floor comprises determining that the ratio exceeds a predetermined threshold ratio, said threshold ratio being at least 0.5, and preferably 0.9.

35. A device for monitoring an object with regard to a fall condition, comprising:

a detector for observing a detection area;

a determination device for determining, based on at least one image from the detector, that an object is lying on a floor in the detection area; and

an alarm device for emitting an alarm a predetermined time period after a determination that an object is lying on a floor by the determination device.

36. The device of claim 35, wherein said determination device further comprises a foreground calculation device for calculating a foreground image, which is the difference between a current image and a predetermined background image;

and calculating the ratio of the foreground image that is present on the floor of the detection area and the total foreground image;

wherein determining that the object is lying on the floor comprises determining that the ratio exceeds a predetermined threshold ratio, said threshold ratio being at least 0.5, and preferably 0.9.

37. The device of claim 35, wherein said determination device further comprises an angle calculation device for calculating an angle between the object and a vertical direction, wherein determining that the object is lying on a floor comprises determining that the angle is above 10 degrees, preferably above 20 degrees.

38. The device of claim 35, wherein said determination device further comprises a length calculation device for determining mass centres of at least two extreme parts of the object;

and calculating a length of the object;

wherein determining that the object is lying on a floor comprises determining that the length of the object is below a predetermined length.

39. The device of claim 35, further comprising of a fall detector for identifying a fall in the detection area, wherein said fall detector is initiated to identify a fall upon said determination device determining that the object is lying on the floor.

40. The device of claims 39, wherein said fall detector comprises:

means for deriving an image sequence for at least a time period preceding the determination, by the determination device, that the object is lying on the floor;

means for analysing the derived image sequence for high velocities and/or negative accelerations; and

means for identifying a fall condition by determining that the velocity is above a predetermined value and/or the acceleration is below a negative value.

41. The device of claim 39, wherein said fall detector comprises;

means for forming a foreground image by calculating a difference between a current image and a previous image;

means for deriving a number of foreground elements from the foreground image; and

means for identifying a fall condition by determining that the number of foreground elements exceeds a foreground number value.