Granted Patent B2

US 11,660,023 · App. 17/632,522 · Granted May 30, 2023

Inventors: Amit Dvash (Zoran, IL); Ronen Tur (Binyamina, IL); Ido Klemer (Givatayim, IL); Alon Keren (Tel-Aviv, IL); Tomer Zimmerman (Tel-Aviv, IL); Michael Orlovsky (Hod Hasharon, IL); Yuval Lomnitz (Herzeliya, IL); Tsachi Rosenhouse (Kiryat Ono, IL)

A61B5/1117A61B5/1126A61B5/746G01S13/66G01S13/88

Abstract

Claims (83)

1. A fall detection system comprising:

a radar unit comprising

at least one transmitter antenna connected to an oscillator and configured to transmit electromagnetic waves into a monitored region,

at least one receiver antenna configured to receive electromagnetic waves reflected by objects within the monitored region and operable to generate raw data,

a processor unit configured to receive raw data from the radar unit and operable to generate fall alerts based upon the received data, and

a communication module configured and operable to communicate the fall alert to third parties,

wherein said processor unit comprises

a data filter configured to receive said raw data, and operable to process the raw data to remove data relating to reflections from static objects thereby generating filtered data,

a tracker module configured to receive the filtered data from the data filter and operable to process the filtered data to identify moving targets and to track the location of the moving targets over time thereby generating target data, and

a fall identification module configured to receive the target data from the tracker module and operable to process the target data by applying fall detection rules and further operable to generate fall alerts, and

wherein said data filter comprises a memory unit, and a microprocessor, and the data filter is operable to

store a first set of raw data set from a first frame in the memory unit,

store a second set of raw data set from a second frame in a memory unit following a selected time interval, and

subtract first frame data from second fame data thereby generating filtered frame data, and

wherein the fall identification module further comprises

a posture detector configured to store target data in a memory unit, to generate an energy profile for each target, and to apply posture selection rules thereby selecting a posture for each target, the posture detector further operable to store a posture history for each target in the memory unit, and

a fall detector configured to access the posture history from the memory unit and to generate a fall alert if at least one target in the monitored region is identified as fallen and no other target in the monitored region is identified as standing.

2. The fall detection system of claim 1 wherein said tracker module comprises a peak detector, an associator and a target tracker wherein:

the peak detector is configured to store filtered frame data, and operable to identity local maxima in each frame thereby generating peak data for the frame;

the associator is configured to store peak data for each frame and to receive tracking data from the target tracker; the associator being operable to associate each peak with a target object, thereby generating target data; and

the target tracker is configured to receive the target data from each frame and operable to populate a target database with a location value and a speed value for each target in each frame, thereby generating tracking data.

3. The fall detection system of claim 1 wherein said processor unit comprises a pre-processor configured and operable to generate energy profiles for target segments of the monitored region and an anomaly detection module configured and operable to identify anomalous energy profiles, the fall detection system further comprising a profile database and wherein the pre-processor comprises an output unit operable to populate the profile database with standard energy profiles and time dependent energy profiles.

4. The fall detection system of claim 3 wherein the anomaly detection module is operable to receive the current energy profile for each target segment from the output unit of the pre-processor and to compare the current energy profile with the corresponding time dependent energy profile stored in the profile database.

5. A fall detection method comprising:

providing at least one radar unit comprising at least one transmitter antenna connected to an oscillator, and at least one receiver antenna configured to receive electromagnetic waves;

providing at least one processor unit configured to receive raw data from the radar unit and operable to generate fall alerts based upon the received data;

providing a communication module configured and operate to communicate a fall alert to third parties;

transmitting electromagnetic waves into a monitored region;

receiving electromagnetic waves reflected from objects in the monitored region;

transferring multiple frames of raw data to processor unit;

removing static objects from the frames of raw data;

identifying moving targets in filtered data;

tracking the moving targets over time;

assigning posture to said targets;

storing a posture history in a memory unit;

applying fall detection rules; and

generating a fall alert if a fall is detected; and

wherein the step of removing static objects from the frames of raw data comprises

collecting raw data from a first frame,

collecting raw data from a second frame, and

subtracting first frame data from the second frame data, and

wherein the step of applying fall detection rules comprises

accessing posture history from the memory unit, and

generating a fall alert only at least one target in the monitored region is identified as fallen and if no other target in the monitored region is identified as standing.

6. The fall detection system of claim 5 wherein the step of identifying moving targets in filtered data comprises:

detecting local maxima within each frame of filtered data;

associating each local maximum with a target object.

7. The fall detection system of claim 5 wherein the step of assigning posture to said targets comprises generating an energy profile for each target by assigning:

a first value for amplitude of reflected energy from an upper region or the target;

a second value for amplitude of reflected energy from a middle region or the target; and

a third value for amplitude of reflected energy from a lower region or the target; and

wherein the step of applying fall detection rules comprises triggering a fall event when any target has an associated third value is higher than both the first value and the second value.

8. The fall detection method of claim 7 further comprising generating a fall alert only if no target is identified having an associated middle value higher than both the first value and the lower value.

9. The fall detection system of claim 5 further comprising:

generating a set of energy profiles;

segmenting the monitored region into target segments;

recording activity within the target region for the duration of a learning period;

for each target segment recording a time dependent profile distribution by recording the frequency of each energy profile during each hour of the learning period; and

populating a profile database with standard energy profiles and time dependent energy profiles.

10. The fall detection system of claim 5 further comprising:

selecting a current profile for each target segment of the monitored region;

communicating the current profile for each target segment to an anomaly detection module;

comparing the current profile for each target segment with the time dependent profile distribution stored in the profile database; and

identifying anomalous energy profiles.

11. The fall detection system of claim 5 further comprising only generating a fall alert if both a fall is detected and an anomalous energy profile is detected.

12. The fall detection system of claim 5 wherein the step of identifying moving targets in filtered data comprises:

setting a peak detection threshold;

detecting local maxima within each frame of filtered data;

defining a stain region for each said local maximum;

selecting peaks by selecting only local maxima having an amplitude above the peak detection threshold and which do not lie within the stain region of a larger local maximum; and

associating each selected peak with a target object.

13. The fall detection method of claim 12 wherein the step of tracking the moving targets over time comprises:

recording a location values for each target in each frame; and

recording a speed values for each target in each frame.

14. The fall detection method of claim 12 wherein the step of tracking the moving targets over time comprises:

recording a location values for each target in each frame;

recording a speed values for each target in each frame;

predicting an expected value for a target in each frame; and

comparing expected value for each target with measured value for each target.

15. The fall detection system of claim 5 wherein the step of assigning posture to said targets comprises:

generating energy profile for each target;

applying posture selection rules;

selecting a current posture.

Continuity (10)