Surveillance system and method for predicting patient falls using motion feature patterns

• US 10,055,961 B1
• Filed: 11/28/2017
• Issued: 08/21/2018
• Est. Priority Date: 07/10/2017
• Status: Active Grant

• Current Assignee: Careview Communications Incorporated
• Original Assignee: Careview Communications Incorporated
• Inventors: Johnson, Steven Gail, del Carpio, Derek
• Primary Examiner(s): Vu, Kim
• Assistant Examiner(s): Delaney, Molly
• Application Number:
  US15/824,552
• Time in Patent Office: 266 Days
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/248 involving reference images or patches
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 30/194 References adjustable by an adaptive method, e.g. learning
• CPC Class Codes: G08B 21/043 detecting an emergency event, e.g. a fall
• CPC Class Codes: G08B 21/0476 Cameras to detect unsafe condition, e.g. video cameras
• CPC Class Codes: G08B 29/186 Fuzzy logic; neural networks

Abstract

A method and system for detecting a fall risk condition, the system comprising a surveillance camera configured to generate a plurality of frames showing an area in which a patient at risk of falling is being monitored, and a computer system comprising memory and logic circuitry configured to store motion feature patterns that are extracted from video recordings, the motion feature patterns are representative of motion associated with real alarm cases and false-alarm cases of fall events, receive a fall alert from a classifier, determine motion features of one or more frames from the plurality of frames that correspond to the fall alert; compare the motion features of the one or more frames with the motion feature patterns, and determine whether to confirm the fall alert based on the comparison.

Claims

1. A surveillance system for detecting a fall risk condition, the system comprising:

a surveillance camera configured to generate a plurality of frames showing an area in which a patient at risk of falling is being monitored; and

a computer system comprising memory and logic circuitry configured to;

store motion feature patterns that are extracted from video recordings, the motion feature patterns are representative of motion associated with real alarm cases and false-alarm cases of fall events;

receive a fall alert from a classifier, the fall alert associated with the plurality of frames generated by the surveillance camera;

determine motion features of one or more frames from the plurality of frames that correspond to the fall alert;

compare the motion features of the one or more frames with the motion feature patterns; and

determine whether to confirm the fall alert based on the comparison.

2. The system of claim 1 wherein configuring the computer system to store the motion feature patterns further comprises the computer system configured to identify the real alarm cases or false-alarm cases of fall events from the video recordings.

3. The system of claim 1 wherein configuring the computer system to compare the motion features with the motion feature patterns further comprises the computer system configured to determine statistically significant similarities between motion features of the fall alert and the stored motion feature patterns.

4. The system of claim 1 wherein configuring the computer system to determine motion features of the one or more frames further comprises the computer system configured to:

detect motion of pixels by comparing pixels of a current frame with at least one previous frame; and

mark pixels that have changed as a motion pixel in a given motion image.

5. The system of claim 1 wherein the computer system is further configured to:

identify the real alarm cases and false-alarm cases of fall events from the plurality of frames; and

determine the motion feature patterns, wherein the motion feature patterns are selected from the group consisting of a centroid, centroid area, connected components ratio, bed motion percentage, and unconnected motion.

6. The system of claim 5 wherein the computer system is further configured to locate the centroid by computing weighted average x and y coordinates of motion pixels in a given motion image.

7. The system of claim 5 wherein the bed motion percentage is a ratio of motion pixels from a given motion image within a virtual bed zone to a total pixel count in the virtual bed zone.

8. The system of claim 5 wherein configuring the computer system to determine the motion feature patterns includes the computer system configured to:

group motion pixels that are connected in a given motion image into clusters; and

prune motion pixels from the given motion image that do not have at least one pixel within a threshold distance of a virtual bed zone.

9. The system of claim 8 wherein the computer system is further configured to determine the connected components ratio based on a ratio of motion pixels outside the virtual bed zone to motion pixels inside the virtual bed zone.

10. The system of claim 8 wherein the computer system is further configured to determine the unconnected motion by calculating an amount of motion pixels in the area of the centroid that is unrelated to connected motion pixels within and near the virtual bed zone.

11. A method for predicting a condition of elevated risk of a fall with a computer system comprising:

storing motion feature patterns that are extracted from video recordings, the motion feature patterns are representative of motion associated with real alarm cases and false-alarm cases of fall events;

receiving a fall alert from a classifier, the fall alert associated with the plurality of frames generated by the surveillance camera;

determining motion features of one or more frames from the plurality of frames that correspond to the fall alert;

comparing the motion features of the one or more frames with the motion feature patterns; and

determining whether to confirm the fall alert based on the comparison.

12. The method of claim 11 wherein storing the motion feature patterns further comprises identifying the real alarm cases or false-alarm cases of fall events from the video recordings.

13. The method of claim 11 wherein comparing the motion features with the motion feature patterns further comprises determining statistically significant similarities between motion features of the fall alert and the stored motion feature patterns.

14. The method of claim 11 wherein determining the motion features of the one or more frames further comprises:

detecting motion of pixels by comparing pixels of a current frame with at least one previous frame; and

marking pixels that have changed as a motion pixel in a given motion image.

15. The method of claim 11 further comprising:

identifying real alarm cases and false-alarm cases from the plurality of frames; and

determining motion feature patterns associated with the real alarm cases and false-alarm cases, the features selected from the group consisting of a centroid, centroid area, connected components ratio, bed motion percentage, and unconnected motion.

16. The method of claim 11 further comprising locating the centroid by computing weighted average x and y coordinates of motion pixels in a given motion image.

17. The method of claim 11 wherein the bed motion percentage is a ratio of motion pixels from a given motion image within a virtual bed zone to a total pixel count in the virtual bed zone.

18. The method of claim 11 wherein determining the motion feature patterns further comprises:

grouping motion pixels that are connected in a given motion image into clusters; and

pruning motion pixels from the given motion image that don'"'"'t have at least one pixel within a threshold distance of the virtual bed zone.

19. The method of claim 18 further comprising determining the connected components ratio based on a ratio of motion pixels outside the virtual bed zone to motion pixels inside the virtual bed zone.

20. The method of claim 18 further comprising determining the unconnected motion by calculating an amount of motion pixels in the area of the centroid that is unrelated to connected motion pixels within and near the virtual bed zone.