Fall detection and reporting technology

Patent metadata

Field	Value
Publication number	US-10037669-B2
Application number	US-201615348550-A
Country	US
Kind code	B2
Filing date	Nov 10, 2016
Priority date	Apr 4, 2011
Publication date	Jul 31, 2018
Grant date	Jul 31, 2018

Abstract

Fall detection and reporting technology, in which output from at least one sensor configured to sense, in a room of a building, activity associated with a patient falling is monitored and a determination is made to capture one or more images of the room based on the monitoring. An image of the room is captured with a camera positioned to include the patient within a field of view of the camera and the captured image of the room is analyzed to detect a state of the patient at a time of capturing the image. A potential fall event for the patient is determined based on the detected state of the patient and a message indicating the potential fall event for the patient is sent based on the determination of the potential fall event for the patient. Techniques are also described for fall detection and reporting using an on-body sensing device.

First claim

What is claimed is:

1. A method comprising:
determining to capture one or more images of an area;
based on the determination to capture one or more images of the area, capturing, with a camera positioned to include a person within a field of view of the camera, an image of the area;
analyzing the captured image of the area to detect a state of the person by:
performing image foreground segmentation on the captured image of the area to create a segmented image,
performing template matching on the segmented image to identify a human shape in the segmented image, and
calculating a position and an orientation associated with the identified human shape in the segmented image; and

determining, based on the calculated position and the calculated orientation associated with the identified human shape in the segmented image, a potential fall event for the person; and
performing an operation directed to assisting the person with the potential fall event based on the determined potential fall event for the person.

2. The method of claim 1 , wherein determining the potential fall event for the person based on the calculated position and the calculated orientation comprises:
monitoring successive image data after calculating the position and the orientation;
comparing the successive image data with prior image data;
determining an activity level of the person based on the comparison of the successive image data with the prior image data; and
classifying the potential fall event based on the determined activity level of the patient.

3. The method of claim 2 , wherein performing the operation directed to assisting the person with the potential fall event comprises handling reporting for the potential fall event based on the classification of the potential fall event.

4. The method of claim 1 , further comprising:
monitoring output from at least one sensor configured to sense, in the area, activity associated with a person falling,
wherein determining to capture one or more images of the area comprises determining to capture one or more images of the area based on the monitoring of output from the at least one sensor.

5. The method of claim 4 :
wherein monitoring output from the at least one sensor comprises:
receiving data indicating a detected change in an orientation of the person and an amount of the orientation change,
receiving data indicating a detected impact and a severity of the detected impact,
determining, based on the received amount of orientation change and the received data indicating the severity of the impact of the person, a threshold for inactivity of the person, and
determining, based on output from the on-body sensor, that the person has been inactive for a period of time greater than the determined threshold; and

wherein determining to capture one or more images of the area comprises determining to capture one or more images of the area based on the determination that the person has been inactive for a period of time greater than the determined threshold.

6. The method of claim 4 :
wherein the at least one sensor configured to sense activity associated with the person falling is a sensor configured to determine a presence of the person in the area; and
wherein monitoring output from the at least one sensor comprises:
receiving, from the sensor configured to determine the presence of the person in the area, a signal indicating that the person is present in the area, and
after a threshold period of time, determining that the person has not left the area and that no further signals have been received from the sensor configured to determine the presence of the person in the area; and

wherein determining to capture one or more images of the room comprises determining to capture one or more images of the room based on determining that the person has not left the area and that no further signals have been received from the sensor configured to determine the presence of the person in the area.

7. The method of claim 4 :
wherein the at least one sensor configured to sense activity associated with the person falling is a button located in the area at a position that permits the person to press the button after a fall; and
wherein monitoring output from the at least one sensor comprises determining that the button has been pressed.

8. The method of claim 1 , wherein performing an operation directed to assisting the person with the potential fall event comprises sending, by a communication device, a message indicating the potential fall event for the person.

9. The method of claim 1 , wherein performing template matching on the segmented image to identify a human shape in the segmented image comprises segmenting a foreground shape falling within an approximate elliptical boundary over a size threshold.

10. The method of claim 1 , wherein performing template matching on the segmented image to identify a human shape in the segmented image comprises identifying at least one non-detection region in the camera's field of view and removing the at least one non-detection region of the camera's field of view from analysis.

11. A system comprising:
at least one processor; and
at least one memory coupled to the at least one processor having stored thereon instructions which, when executed by the at least one processor, causes the at least one processor to perform operations comprising:
determining to capture one or more images of an area;
based on the determination to capture one or more images of the area, capturing, with a camera positioned to include a person within a field of view of the camera, an image of the area;
analyzing the captured image of the area to detect a state of the person by:
performing image foreground segmentation on the captured image of the area to create a segmented image,
performing template matching on the segmented image to identify a human shape in the segmented image, and
calculating a position and an orientation associated with the identified human shape in the segmented image; and

determining, based on the calculated position and the calculated orientation associated with the identified human shape in the segmented image, a potential fall event for the person; and
performing an operation directed to assisting the person with the potential fall event based on the determined potential fall event for the person.

12. The system of claim 11 , wherein determining the potential fall event for the person based on the calculated position and the calculated orientation comprises:
monitoring successive image data after calculating the position and the orientation;
comparing the successive image data with prior image data;
determining an activity level of the person based on the comparison of the successive image data with the prior image data; and
classifying the potential fall event based on the determined activity level of the patient.

13. The system of claim 12 , wherein performing the operation directed to assisting the person with the potential fall event comprises handling reporting for the potential fall event based on the classification of the potential fall event.

14. The system of claim 11 , wherein the operations further comprise:
monitoring output from at least one sensor configured to sense, in the area, activity associated with a person falling, and
wherein determining to capture one or more images of the area comprises determining to capture one or more images of the area based on the monitoring of output from the at least one sensor.

15. The system of claim 14 :
wherein monitoring output from the at least one sensor comprises

Assignees

• Alarm Com Inc

Inventors

• Hanson Mark Andrew
• Martin Jean-Paul
• Barth Adam T
• Silverman Christopher

Classifications

• G08B21/02

  Alarms for ensuring the safety of persons · CPC title

• A61B5/0022

  Monitoring a patient using a global network, e.g. telephone networks, internet · CPC title

• G08B21/043 Primary

  detecting an emergency event, e.g. a fall · CPC title

• A61B5/0077

  Devices for viewing the surface of the body, e.g. camera, magnifying lens · CPC title

• G06T7/0042

  Physics · mapped topic