Detecting user activity based on location data

• US 10,451,744 B2
• Filed: 06/02/2017
• Issued: 10/22/2019
• Est. Priority Date: 06/02/2017
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: Wang, Lei, Miller, Isaac Thomas, Bae, Hyojoon, Selle, Christina, MacGougan, Glenn Donald
• Primary Examiner(s): Shin, Soo
• Application Number:
  US15/613,039
• Publication Number:
  US 20180349728A1
• Time in Patent Office: 872 Days
• Field of Search: None
• CPC Class Codes: G01S 19/19 Sporting applications
• CPC Class Codes: G01S 19/485 whereby the further system is an optical system or imaging system
• CPC Class Codes: G01S 5/017 Detecting state or type of motion
• CPC Class Codes: G01S 5/0294 Trajectory determination or predictive filtering, e.g. target tracking or Kalman filtering
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 7/248 involving reference images or patches
• CPC Class Codes: G06T 7/277 involving stochastic approaches, e.g. using Kalman filters
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
• CPC Class Codes: G06V 40/23 Recognition of whole body movements, e.g. for sport training

Abstract

The disclosed embodiments are directed to detecting a user activity based on patterns in location data. In an embodiment, a method comprises: obtaining, by a processor of a computing device, location data; detecting, by the processor, a pattern in the location data; determining, by the processor and based on the detected pattern, an activity associated with a mobile device; and adjusting, by the processor, one or more parameters of a location data filter configured to process the location data.

Claims

1. A method comprising:

obtaining, by a processor of a computing device, estimated position data indicating an activity path of the computing device, the estimated position data generated by a location data filter of the computing device;

detecting, by the processor, a pattern in the estimated position data;

determining, by the processor and based on the detected pattern, an activity associated with the activity path;

based on the determined activity, adjusting, by the processor, one or more parameters of the location data filter; and

generating, by the location data filter, the estimated position data, wherein the estimated position data is smoothed by the location data filter by an amount determined by the adjusted one or more parameters.

2. The method of claim 1, wherein detecting a pattern in the estimated position data comprises:

detecting a loop in the estimated position data.

3. The method of claim 2, wherein detecting a loop in the estimated position data comprises:

stepping through a sequence of the estimated position data at different rates; and

comparing, at each step in the sequence, the estimated position data that corresponds to respective locations in the sequence using an equivalence threshold.

4. The method of claim 3, wherein the equivalence threshold is determined at least by estimated position, speed and estimated position uncertainty.

5. The method of claim 2, wherein detecting the loop in the estimated position data further comprises:

determining a center or centroid of a subset of the estimated position data included in the loop;

calculating a distance of each estimated position data in the subset of estimated position data from the center or centroid;

determining a percentage of the subset of the estimated position data that falls within a minimum and maximum boundary determined by dimensions of a reference running track; and

determining that the percentage of the estimated position data that falls within the minimum and maximum boundaries exceeds a specified threshold value.

6. The method of claim 1, wherein the location data filter is a Bayesian filter and the one or more parameters are process noise.

7. A method comprising:

generating, by one or more processors of a computing device, an image of estimated position data indicating an activity path of the computing device;

determining, by the one or more processors, a contour in the image, the contour representing the activity path;

determining, by the one or more processors, a path descriptor for the activity path;

matching, by the one or more processors, the path descriptor with one of a plurality of reference path descriptors;

responsive to the matching, adjusting, by the one or more processors, one or more parameters of a location data filter used to generate the estimated position data; and

generating, by the location data filter, the estimated position data, wherein the estimated position data is smoothed by the location data filter by an amount determined by the adjusted one or more parameters.

8. The method of claim 7, wherein the path descriptor and reference path descriptors are feature vectors that include a plurality of image moment invariants.

9. The method of claim 7, wherein the image is morphed before determining the contour in the image.

10. A system comprising:

one or more processors;

memory coupled to the one or more processors and configured to store instructions, which, when executed by the one or more processors, causes the one or more processors to perform operations comprising;

obtaining, from a location data filter, estimated position data indicating an activity path of a computing device, the estimated position data generated by a location data filter of the computing device;

detecting a pattern in the estimated position data;

determining an activity associated with the activity path based on the detected pattern;

based on the determined activity, adjusting one or more parameters of the location data filter; and

generating, by the location data filter, the estimated position data, wherein the estimated position data is smoothed by the location data filter by an amount determined by the adjusted one or more parameters.

11. The system of claim 10, wherein detecting a pattern in the estimated position data comprises:

detecting a loop in the estimated position data.

12. The system of claim 11, wherein detecting a loop in the estimated position data comprises:

stepping through a sequence of the estimated position data at different rates; and

comparing, at each step in the sequence, the estimated position data that corresponds to respective locations in the sequence using an equivalence threshold.

13. The system of claim 12, wherein the equivalence threshold is determined at least by estimated position, speed and estimated position uncertainty.

14. The system of claim 11, wherein detecting the loop in the estimated position data further comprises:

determining a center or centroid of a subset of the estimated position data included in the loop;

calculating a distance of each estimated position data in the subset of the estimated position data from the center or centroid;

determining a percentage of the subset of the estimated position data that falls within a minimum and maximum boundary determined by dimensions of a reference running track; and

determining that the percentage of the estimated position data that falls within the minimum and maximum boundaries exceeds a specified threshold value.

15. The system of claim 10, wherein the location data filter is a Bayesian filter and the one or more parameters are process noise.

16. A system comprising:

one or more processors;

memory coupled to the one or more processors and configured to store instructions, which, when executed by the one or more processors, cause the one or more processors to perform operations comprising;

generating an image of estimated position data indicating an activity path of a computing device, the estimated position data generated by a location data filter of the computing device;

determining a contour in the image, the contour representing the activity path;

determining a path descriptor for the activity path;

matching the path descriptor with one of a plurality of reference path descriptors;

responsive to the matching, adjusting, by the one or more processors, one or more parameters of the location data filter; and

generating, by the location data filter, the estimated position data, wherein the estimated position data is smoothed by the location data filter by an amount determined by the adjusted one or more parameters.

17. The system of claim 16, wherein the path descriptor and reference path descriptors are feature vectors that include a plurality of image moment invariants.

18. The system of claim 16, wherein the image is morphed before determining the contour in the image.

19. A method comprising:

obtaining, by a processor of a computing device, estimated position data indicating an activity path of the computing device, the estimated position data generated by a location data filter of the computing device;

determining, by the processor, whether the activity path includes a loop;

in accordance with determining that the activity path includes a loop, adjusting, by the processor, one or more parameters of the location data filter; and

generating, by the location data filter, the estimated position data, wherein the estimated position data is smoothed by the location data filter by an amount determined by the adjusted one or more parameters.

20. The method of claim 19, wherein determining whether the activity path includes a loop, further comprises:

determining a center or centroid of a subset of the estimated position data included in the activity path;

calculating a distance of each estimated position data in the subset of estimated position data from the center or centroid;

determining a percentage of the subset of the estimated position data that falls within a minimum and maximum boundary determined by dimensions of a reference activity path; and

determining that the percentage of the estimated position data that falls within the minimum and maximum boundaries exceeds a specified threshold value.

21. A method comprising:

generating, by one or more processors of a computing device, an image of estimated position data indicating an activity path of the computing device, the estimated position data generated by a location data filter of the computing device;

determining, by the one or more processors, whether the activity path includes a loop;

in accordance with determining that the activity path includes a loop;

determining, by the one or more processors, a path descriptor for the activity path;

matching, by the one or more processors, the path descriptor with one of a plurality of reference path descriptors;

responsive to the matching, adjusting, by the one or more processors, one or more parameters of the location data filter; and

generating, by the location data filter, the estimated position data, wherein the estimated position data is smoothed by the location data filter by an amount determined by the adjusted one or more parameters.

22. The method of claim 21, wherein determining whether the activity path includes a loop, further comprises:

determining a center or centroid of a subset of the estimated position data included in the activity path;

calculating a distance of each estimated position data in the subset of estimated position data from the center or centroid;

determining a percentage of the subset of the estimated position data that falls within a minimum and maximum boundary determined by dimensions of a reference activity path; and

determining that the percentage of the estimated position data that falls within the minimum and maximum boundaries exceeds a specified threshold value.