Methods and devices for identifying users based on tremor

• US 8,994,657 B2
• Filed: 06/20/2007
• Issued: 03/31/2015
• Est. Priority Date: 04/30/2004
• Status: Expired due to Fees

• Current Assignee: DRNC Holdings, Inc. (InterDigital, Inc.)
• Original Assignee: Hillcrest Laboratories Incorporated (CEVA Incorporated)
• Inventors: Liberty, Matthew G., Simpkins, Daniel S., Gritton, Charles W. K., Roller, Christopher D.
• Primary Examiner(s): Sitta, Grant
• Application Number:
  US11/820,515
• Publication Number:
  US 20070247425A1
• Time in Patent Office: 2,841 Days
• Field of Search: 345156-158
• US Class Current: 345/158
• CPC Class Codes: A61B 5/1101 Detecting tremor
• CPC Class Codes: A61B 5/117 Identification of persons methods or arrangements for recognising patterns, e.g. fingerprints, G06F18/00, G06V40/00; identification of persons by analysing their voice or speech G10L17/00
• CPC Class Codes: A61B 5/1171 based on the shapes or appearances of their bodies or parts thereof
• CPC Class Codes: A61B 5/6825 Hand
• CPC Class Codes: A61B 5/6897 Computer input devices, e.g. mice or keyboards
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 3/017 Gesture based interaction, e.g. based on a set of recognized hand gestures interaction based on gestures traced on a digitiser G06F3/04883
• CPC Class Codes: G06F 3/0346 with detection of the device orientation or free movement in a 3D space, e.g. 3D mice, 6-DOF [six degrees of freedom] pointers using gyroscopes, accelerometers or tilt-sensors
• CPC Class Codes: G06F 3/038 Control and interface arrangements therefor, e.g. drivers or device-embedded control circuitry
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G08C 2201/32 Remote control based on movements, attitude of remote control device
• CPC Class Codes: G08C 2201/61 Password, biometric
• CPC Class Codes: H04N 21/42204 User interfaces specially adapted for controlling a client device through a remote control device; Remote control devices therefor constructive details of casings for the remote control device H01H9/0235; user interfaces for controlling a tuning device of a television receiver through a remote control H03J9/00; remote control of peripheral devices connected to a television receiver through the remote control device of the television receiver H04B1/205
• CPC Class Codes: H04N 21/42222 Additional components integrated in the remote control device, e.g. timer, speaker, sensors for detecting position, direction or movement of the remote control, microphone or battery charging device

Abstract

Systems and methods according to the present invention address these needs and others by providing a handheld device, e.g., a 3D pointing device, which uses hand tremor as an input. One or more sensors within the handheld device detect a user'"'"'s hand tremor and identify the user based on the detected tremor.

Claims

1. A method for identifying a user of a handheld device comprising the steps of:

detecting, using a motion sensor and a processor, a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said step of identifying further comprises;

(a) obtaining tremor data sets associated with a plurality of users;

(b) extracting features from each of said tremor data sets to generate an extracted feature set for each of said plurality of users,

(c) removing features from each of said extracted feature sets to generate a reduced feature set for each of said plurality of users,

(d) identifying clusters associated with said reduced feature sets, and

(e) identifying said user by generating new feature vectors based on said hand tremor data and determining whether said new features lie within said clusters,

wherein said step (b) of extracting features further comprises;

calculating a plurality of point Fast Fourier Transforms (FFTs) averaged with an overlap for a number of points in said movement data within a predetermined frequency range.

2. The method of claim 1, further comprising the step of:

selectively turning on a system associated with said handheld device on based upon said step of identifying.

3. The method of claim 1, further comprising the step of:

restricting access to a system associated with said handheld device based upon an identity of said user.

4. The method of claim 3, wherein said system is a media system and said step of restricting access further comprises the step of:

selectively permitting said user to access a media item based upon said identifying step.

5. The method of claim 1, further comprising the step of:

determining an identity of said user by comparing a set of detected tremor features with a plurality of stored tremor classes.

6. The method of claim 5, wherein said detected tremor features and said plurality of stored tremor classes are associated with frequency domain data.

7. The method of claim 5, wherein said detected tremor features and said plurality of stored tremor classes are associated with time domain data.

8. The method of claim 5, wherein if a match does not occur between said detected tremor features and said plurality of stored tremor classes, then said detected tremor features are used to establish a new user identity.

9. The method of claim 5, wherein if a match does not occur between said tremor features and said plurality of stored tremor classes, then said user is assigned a default set of system access permissions.

10. The method of claim 1, further comprising the step of:

retrieving preference settings associated with said user in response to said identifying step.

35. The method of claim 1, wherein said step of identifying further comprises:

generating feature vectors associated with said hand tremor data and determining in which of a plurality of clusters said feature vectors lie.

36. The method of claim 1, wherein said step (a) of obtaining tremor data further comprises:

performing a training process wherein each of said plurality of users holds said handheld device without introducing intentional motion for a predetermined period of time.

37. The method of claim 1, wherein said step (a) of obtaining tremor data further comprises:

gathering said tremor data as each of said plurality of users uses said handheld device.

38. The method of claim 1, wherein said step (b) of extracting features further comprises:

using low frequency spectra from a power spectral density of the movement data.

39. The method of claim 1, wherein said step (c) of removing features further comprises:

removing features from said extracted feature set which are less useful in differentiating between said plurality of users.

40. The method of claim 39, wherein said removed features include DC values associated with said movement data.

41. The method of claim 39, wherein said reduced feature set includes 9 Hz values associated with said movement data.

42. The method of claim 1, wherein said step (c) of removing features further comprises:

applying a Principal Component Analysis (PCA) algorithm to determine a set of basis vectors.

43. The method of claim 1, wherein said step (d) of identifying clusters further comprises:

employing K-means clustering or RBF neural net classification to generate said clusters.

44. The method of claim 1, wherein said step (d) of identifying clusters further comprises:

applying a discriminant to accentuate at least one discriminating feature associated with each cluster.

45. The method of claim 44, wherein said discriminant is an Enhanced Fisher Linear Discriminant (EFM-1).

11. A non-transitory computer-readable medium, capable of storing program instructions which, when executed perform the steps of:

detecting a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said step of identifying further comprises;

(a) obtaining tremor data sets associated with a plurality of users;

(b) extracting features from each of said tremor data sets to generate an extracted feature set for each of said plurality of users,

(c) removing features from each of said extracted feature sets to generate a reduced feature set for each of said plurality of users,

(d) identifying clusters associated with said reduced feature sets, and

(e) identifying said user by generating new feature vectors based on said hand tremor data and determining whether said new features lie within said clusters,

wherein said step (b) of extracting features further comprises;

calculating a plurality of point Fast Fourier Transforms (FFTs) averaged with an overlap for a number of points in said movement data within a predetermined frequency range.

12. The non-transitory computer-readable medium of claim 11, further comprising the step of:

selectively turning on a system associated with said handheld device on based upon said step of identifying.

13. The non-transitory computer-readable medium of claim 11, further comprising the step of:

restricting access to a system associated with said handheld device based upon an identity of said user.

14. The non-transitory computer-readable medium of claim 13, wherein said system is a media system and said step of restricting access further comprises the step of:

selectively permitting said user to access a media item based upon said identifying step.

15. The non-transitory computer-readable medium of claim 11, further comprising the step of:

determining an identity of said user by comparing a set of detected tremor features with a plurality of stored tremor classes.

16. The non-transitory computer-readable medium of claim 15, wherein said detected tremor features and said plurality of stored tremor classes are associated with frequency domain data.

17. The non-transitory computer-readable medium of claim 15, wherein said detected tremor features and said plurality of stored tremor classes are associated with time domain data.

18. The non-transitory computer-readable medium of claim 15, wherein if a match does not occur between said detected tremor features and said plurality of stored tremor classes, then said detected tremor features are used to establish a new user identity.

19. The non-transitory computer-readable medium of claim 15, wherein if a match does not occur between said tremor features and said plurality of stored tremor classes, then said user is assigned a default set of system access permissions.

20. The non-transitory computer-readable medium of claim 11, further comprising the step of:

retrieving preference settings associated with said user in response to said identifying step.

46. The non-transitory computer-readable medium of claim 11, wherein said step of identifying further comprises:

generating feature vectors associated with said hand tremor data and determining in which of a plurality of clusters said feature vectors lie.

47. The non-transitory computer-readable medium of claim 11, wherein said step (a) of obtaining tremor data further comprises:

performing a training process wherein each of said plurality of users holds said handheld device without introducing intentional motion for a predetermined period of time.

48. The non-transitory computer-readable medium of claim 11, wherein said step (a) of obtaining tremor data further comprises:

gathering said tremor data as each of said plurality of users uses said handheld device.

49. The non-transitory computer-readable medium of claim 11, wherein said step (b) of extracting features further comprises:

using low frequency spectra from a power spectral density of the movement data.

50. The non-transitory computer-readable medium of claim 11, wherein said step (c) of removing features further comprises:

removing features from said extracted feature set which are less useful in differentiating between said plurality of users.

51. The non-transitory computer-readable medium of claim 50, wherein said removed features include DC values associated with said movement data.

52. The non-transitory computer-readable medium of claim 50, wherein said reduced feature set includes 9 Hz values associated with said movement data.

53. The non-transitory computer-readable medium of claim 11, wherein said step (c) of removing features further comprises:

applying a Principal Component Analysis (PCA) algorithm to determine a set of basis vectors.

54. The non-transitory computer-readable medium of claim 11, wherein said step (d) of identifying clusters further comprises:

employing K-means clustering or RBF neural net classification to generate said clusters.

55. The non-transitory computer-readable medium of claim 11, wherein said step (d) of identifying clusters further comprises:

applying a discriminant to accentuate at least one discriminating feature associated with each cluster.

56. The non-transitory computer-readable medium of claim 55, wherein said discriminant is an Enhanced Fisher Linear Discriminant (EFM-1).

21. A system including a handheld device, the system comprising:

at least one motion sensor capable of generating data associated with movement of the handheld device; and

a processor for generating hand tremor data based on said movement data and for identifying a user of said handheld device based on said hand tremor data by determining within which of a plurality of clusters at least one feature vector associated with said hand tremor data lies.

22. The system of claim 21, wherein said at least one motion sensor includes at least one of:

a rotational sensor, an inertial sensor, an accelerometer, a gyroscope, a magnetic sensor, an optical sensor and a camera.

23. The system of claim 21, wherein said processor is configured to identify said user by:

(a) obtaining tremor data sets associated with a plurality of users;

(b) extracting features from each of said tremor data sets to generate an extracted feature set for each of said plurality of users,

(c) removing features from each of said extracted feature sets to generate a reduced feature set for each of said plurality of users,

(d) identifying clusters associated with said reduced feature sets, and

(e) identifying said user by generating new feature vectors based on said hand tremor data and determining whether said new features lie within said clusters.

24. The system of claim 23, wherein said step (a) of obtaining tremor data further comprises:

performing a training process wherein each of said plurality of users holds said handheld device without introducing intentional motion for a predetermined period of time.

25. The system of claim 23, wherein said step (a) of obtaining tremor data further comprises:

gathering said tremor data as each of said plurality of users uses said handheld device.

26. The system of claim 23, wherein said step (b) of extracting features further comprises:

calculating a plurality of point Fast Fourier Transforms (FFTs) averaged with an overlap for a number of points in said movement data within a predetermined frequency range.

27. The system of claim 23, wherein said step (b) of extracting features further comprises:

using low frequency spectra from a power spectral density of the movement data.

28. The system of claim 23, wherein said step (c) of removing features further comprises:

removing features from said extracted feature set which are less useful in differentiating between said plurality of users.

29. The system of claim 28, wherein said removed features include DC values associated with said movement data.

30. The system of claim 28, wherein said reduced feature set includes 9 Hz values associated with said movement data.

31. The system of claim 23, wherein said step (c) of removing features further comprises:

applying a Principal Component Analysis (PCA) algorithm to determine a set of basis vectors.

32. The system of claim 23, wherein said step (d) of identifying clusters further comprises:

employing K-means clustering or RBF neural net classification to generate said clusters.

33. The system of claim 23, wherein said step (d) of identifying clusters further comprises:

applying a discriminant to accentuate at least one discriminating feature associated with each cluster.

34. The system of claim 33, wherein said discriminant is an Enhanced Fisher Linear Discriminant (EFM-1).

57. A method for identifying a user of a handheld device comprising the steps of:

detecting, using a motion sensor and a processor, a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said step of identifying further comprises;

(a) obtaining tremor data sets associated with a plurality of users;

(b) extracting features from each of said tremor data sets to generate an extracted feature set for each of said plurality of users,

(c) removing features from each of said extracted feature sets to generate a reduced feature set for each of said plurality of users,

(d) identifying clusters associated with said reduced feature sets, and

(e) identifying said user by generating new feature vectors based on said hand tremor data and determining whether said new features lie within said clusters,

wherein said step (c) of removing features further comprises;

applying a Principal Component Analysis (PCA) algorithm to determine a set of basis vectors.

58. A method for identifying a user of a handheld device comprising the steps of:

detecting, using a motion sensor and a processor, a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said step of identifying further comprises;

(a) obtaining tremor data sets associated with a plurality of users;

(b) extracting features from each of said tremor data sets to generate an extracted feature set for each of said plurality of users,

(c) removing features from each of said extracted feature sets to generate a reduced feature set for each of said plurality of users,

(d) identifying clusters associated with said reduced feature sets, and

(e) identifying said user by generating new feature vectors based on said hand tremor data and determining whether said new features lie within said clusters,

wherein said step (d) of identifying clusters further comprises;

applying a discriminant to accentuate at least one discriminating feature associated with each cluster,

wherein said discriminant is an Enhanced Fisher Linear Discriminant (EFM-1).

59. A non-transitory computer-readable medium, capable of storing program instructions which, when executed perform the steps of:

detecting a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said step of identifying further comprises;

(a) obtaining tremor data sets associated with a plurality of users;

(b) extracting features from each of said tremor data sets to generate an extracted feature set for each of said plurality of users,

(c) removing features from each of said extracted feature sets to generate a reduced feature set for each of said plurality of users,

(d) identifying clusters associated with said reduced feature sets, and

(e) identifying said user by generating new feature vectors based on said hand tremor data and determining whether said new features lie within said clusters,

wherein said step (c) of removing features further comprises;

applying a Principal Component Analysis (PCA) algorithm to determine a set of basis vectors.

60. A non-transitory computer-readable medium, capable of storing program instructions which, when executed perform the steps of:

detecting a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said step of identifying further comprises;

(a) obtaining tremor data sets associated with a plurality of users;

(b) extracting features from each of said tremor data sets to generate an extracted feature set for each of said plurality of users,

(c) removing features from each of said extracted feature sets to generate a reduced feature set for each of said plurality of users,

(d) identifying clusters associated with said reduced feature sets, and

(e) identifying said user by generating new feature vectors based on said hand tremor data and determining whether said new features lie within said clusters,

wherein said step (d) of identifying clusters further comprises;

applying a discriminant to accentuate at least one discriminating feature associated with each cluster,

wherein said discriminant is an Enhanced Fisher Linear Discriminant (EFM-1).

61. A method for identifying a user of a handheld device comprising the steps of:

detecting, using a motion sensor and a processor, a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said motion sensor is capable of generating data associated with movement of the handheld device,

and said processor generates hand tremor data based on said movement data for identifying said user of said handheld device based on said hand tremor data by determining within which a plurality of clusters at least one feature vector associated with said hand tremor data lies.

62. A non-transitory computer readable medium, capable of storing program instructions which, when executed perform the steps of:

detecting a hand tremor associated with a user holding said handheld device; and

identifying said user based on said detected hand tremor,

wherein said detecting step is performed by at least one motion sensor capable of generating data associated with movement of the handheld device, and

wherein said identifying step is performed by at least one processor for generating hand tremor data based on said movement data and for identifying a user of said handheld device based on said hand tremor data by determining within which of a plurality of clusters at least one feature vector associated with said hand tremor data lies.