Use of thermopiles to detect human location

• US 10,302,810 B2
• Filed: 01/30/2017
• Issued: 05/28/2019
• Est. Priority Date: 01/30/2017
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Grosse-Puppendahl, Tobias, Scott, James, Sreedhar, Rinku, Helmes, John Franciscus Marie, Li, Hanchuan
• Primary Examiner(s): Porta, David P
• Assistant Examiner(s): Valentiner, Jeremy S
• Application Number:
  US15/420,006
• Publication Number:
  US 20180217292A1
• Time in Patent Office: 848 Days
• CPC Class Codes: A61B 5/01 Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue clinical contact thermometers G01K13/20
• CPC Class Codes: G01J 5/0025 Living bodies ear thermometers G01J5/0011; detecting, measuring or recording for diagnostic purposes A61B5/00
• CPC Class Codes: G01J 5/12 using thermoelectric elements, e.g. thermocouples
• CPC Class Codes: G01J 5/80 Calibration using comparison with reference sources G01J5/52
• CPC Class Codes: G01V 8/20 using multiple transmitters or receivers
• CPC Class Codes: G06N 5/046 Forward inferencing; Production systems

Abstract

A method of detecting presence and location uses sensor data received from a plurality of thermopiles, each thermopile having a different field of view. In response to detecting a change in the sensor data, stored background values for each field of view are accessed and then the location of a body (e.g. a human or animal) is determined based on differences between the sensor data and sensor values predicted using a forward model and the stored background values for each field of view. Having determined the location, the stored background values are updated based on differences between the sensor data and the predicted sensor values for a body at the determined location.

Claims

1. A method of detecting presence and location of a body using a mobile computing device, the method comprising:

receiving sensor data from a plurality of thermopiles coupled to the mobile device, each thermopile having a different field of view;

in response to detecting a change in the sensor data, accessing stored background values for each field of view;

determining, using a forward model, sensor values corresponding to locations of the body by applying the sensor data to a lookup table within the forward model;

determining a location of the body based on differences between sensor values within the sensor data and the determined sensor values from the forward model, and the differences between the sensor data and the stored background values for each field of view; and

updating the stored background values based on differences between the sensor data and the determined sensor values.

2. The method according to claim 1, wherein the location is determined by reducing the differences between the sensor data and sensor values predicted using the forward model and the stored background values for each field of view.

3. The method according to claim 1, wherein the stored background values are updated by changing the stored background values to reduce the differences between the sensor data and sensor values predicted using the forward model and the stored background values for each field of view for the body at the determined location.

4. The method according to claim 3, wherein the stored background values are updated by changing the stored background values to reduce the differences between the sensor data and sensor values predicted using the forward model and the stored background values for each field of view for the body at the determined location.

5. The method according to claim 1, wherein the field of view of one of the plurality of thermopiles partially overlaps the field of view of a different one of the plurality of thermopiles.

6. The method according to claim 1, wherein the location of the body comprises an angular position of the body relative to the plurality of thermopiles or an indication that no body is present.

7. The method according to claim 6, wherein in response to determining that the location comprises an indication that no body is present, updating the stored background values comprises:

determining if there is a correlation between the differences between the sensor data and the predicted sensor values for each of the plurality of thermopiles and in response to identifying a correlation, updating the stored background values to be equal to the sensor data.

8. The method according to claim 6, wherein in response to determining that the location comprises an indication that no body is present, updating the stored background values comprises:

in response to determining that no body has been present for more than a predefined period of time, updating the stored background values to be equal to the sensor data.

9. The method according to claim 6, wherein the location of the body further comprises a distance of the body from the plurality of thermopiles.

10. The method according to claim 1, further comprising:

storing an error correction value for each of the plurality of thermopiles;

resetting the error correction values to be equal to the stored background values each time the stored background values are updated;

incrementing the error correction values over time and/or in correspondence to features extracted from sensor data for two or more of the plurality of thermopiles; and

in response to determining that the error correction value for any of the plurality of thermopiles equals the sensor data for that thermopile, updating the stored background value for the thermopile to be equal to the sensor data for that thermopile.

11. The method according to claim 1, further comprising:

in response to detecting a change in the sensor data, determining if the body is present;

and wherein the method comprises;

in response to determining that the body is present, determining a location of the body based on differences between the sensor data and sensor values predicted using a forward model and the stored background values for each field of view; and

updating the stored background values based on differences between the sensor data and the predicted sensor values; and

in response to determining that the body is not present, updating the stored background values based on the sensor data.

12. A system for detecting presence and location of a body using a mobile computing device, the system comprising:

a plurality of thermopiles coupled to the mobile computing device, each thermopile having a different field of view;

a location detection module;

a forward model; and

a data store arranged to store a background value for each of the plurality of thermopiles,

wherein the location detection module is arranged to receive sensor data from the plurality of thermopiles and in response to detecting a change in the sensor data, to access the stored background values for each field of view, determine using a forward model, sensor values corresponding to locations of the body by applying the sensor data to a lookup table within the forward model, determine a location of the body based on differences between sensor values within the sensor data and the determined sensor values from the forward model, and the differences between the sensor data and the stored background values; and

to update the stored background values based on differences between the sensor data and the determined sensor values.

13. The system according to claim 12, wherein the location of the body is determined by reducing the differences between the sensor data and sensor values predicted using the forward model and the stored background values for each field of view.

14. The system according to claim 12, wherein the stored background values are updated by changing the stored background values to reduce the differences between the sensor data and sensor values predicted using the forward model and the stored background values for each field of view for the body at the determined location.

15. The system according to claim 12, wherein the system further comprises a motion sensor and wherein the location detection module is further arranged to update the stored background values in response to motion sensor data.

16. The system according to claim 12, further comprising a plurality of lenses and wherein the lenses are positioned in front of the plurality of thermopiles.

17. The system according to claim 12, further comprising a display device and wherein the plurality of thermopiles are integrated into or mounted on the display device.

18. The system according to claim 12, wherein the forward model comprises a lookup table arranged to store a value for a maximum forward contribution of the body to the sensor values and a plurality of values detailing the effect of the foreground contribution of the body at different locations relative to the plurality of thermopiles.

19. The system as claimed in claim 12, the location detection module being at least partially implemented using hardware logic selected from any one or more of:

a field-programmable gate array, a program-specific integrated circuit, a program-specific standard product, a system-on-a-chip, a complex programmable logic device.

20. A computing device for detecting presence and location of a body, the computing device comprising:

an input for receiving sensor data from a plurality of thermopiles coupled to the computing device, each thermopile having a different field of view;

a processor; and

memory configured to store background values for each of the plurality of thermopiles, data defining a forward model and computer executable instructions, which when executed cause the processor, in response to detecting a change in the sensor data, to;

access the stored background values for each field of view;

determine, using a forward model, sensor values corresponding to locations of the body by applying the sensor data to a lookup table within the forward model;

determine a location of the body based on differences between sensor values within the sensor data and the determined sensor values from the the forward model, and the differences between the sensor data and the stored background values; and

update the stored background values based on differences between the sensor data and the determined sensor values.