Security System Utilizing Gesture Recognition

• US 20100052851A1
• Filed: 09/04/2008
• Published: 03/04/2010
• Est. Priority Date: 09/04/2008
• Status: Active Grant

• Current Assignee: Northrop Grumman Systems Corporation (Northrop Grumman Corporation)
• Original Assignee: Northrop Grumman Systems Corporation (Northrop Grumman Corporation)
• Inventors: Kaehler, Adrian
• Granted Patent: US 9,251,407 B2
• Time in Patent Office: Days
• US Class Current: 340/5.810
• CPC Class Codes: G06F 21/316 by observing the pattern of computer usage, e.g. typical user behaviour
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 21/83 input devices, e.g. keyboards, mice or controllers thereof
• CPC Class Codes: G06F 2221/2101 Auditing as a secondary aspect
• 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 3/0421 by interrupting or reflecting a light beam, e.g. optical touch-screen
• CPC Class Codes: G06F 3/04883 for inputting data by handwriting, e.g. gesture or text
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16
• CPC Class Codes: G06V 40/30 Writer recognition; Reading and verifying signatures
• CPC Class Codes: H04L 63/0861 using biometrical features, e.g. fingerprint, retina-scan cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231

Abstract

Systems and methods are provided for restricting access to an item of interest. A normalization component resamples an input trajectory to produce a resampled trajectory having a standard size. A reference point generator reduces the resampled trajectory to respective values for a set of reference points, each having at least two associated coordinates. The system further includes at least one authentication region. Each of the at least one authentication region represents at least one of the set of reference points. A verification component is configured to determine if the values for the set of reference points from a given input falls within the at least one authentication region. An access restriction mechanism restricts access to the item of interest unless a threshold number of values for the set of reference points from the input falls within their associated authentication regions.

Claims

1. A security system for restricting access to an item of interest, comprising:

a gesture input interface configured to convert a continuous gesture from a user into an input trajectory;

a normalization component configured to resample the input trajectory to produce a resampled trajectory having a standard size;

a reference point generator configured to reduce the resampled trajectory to respective values for a set of reference points, each of the values for the set of reference points having at least two associated coordinates;

at least one authentication region, each of the at least one authentication region representing at least one of the set of reference points, such that each authentication region is defined within a multidimensional feature space having dimensions corresponding to the at least two coordinates associated with each of the at least one points associated with the authentication region from respective sets of representative statistics for the at least one reference point represented by the authentication region;

a verification component configured to determine if the values for the set of reference points from a given input falls within the at least one authentication region; and

an access restriction mechanism configured to restrict access to the item of interest unless a threshold number of values for the set of reference points from the input falls within their associated authentication regions.

2. The system of claim 1, wherein the at least one authentication region comprises only one authentication region defined within a multidimensional feature space having dimensions corresponding to the at least two coordinates associated with each point, the verification component being configured to determine if a feature vector representing a given input trajectory falls within the authentication region.

3. The system of claim 1, wherein the at least one authentication region comprises an associated authentication region for each of the set of reference points, each authentication region being defined within a multidimensional feature space having dimensions corresponding to the at least two coordinates associated with each point, the verification component being configured to determine if respective feature vectors representing each reference point from a resampled trajectory falls within its associated authentication region.

4. The system of claim 1, further comprising a parameter calculation component configured to receive training data representing a characteristic gesture for the user, comprising values for the set of reference points derived from user input at the gesture input interface, the normalization component, and the reference point generator, and calculate the set of representative statistics for each of the set of reference points, the representative statistics comprising a mean value for each of the at least two associated coordinates, a variance value for each of the at least two coordinates, and a covariance between a first coordinate and a second coordinate of the at least two coordinates.

5. The system of claim 1, the item of interest comprising a credit card account.

6. The system of claim 1, the access restriction mechanism comprising a set of executable instructions, stored on a computer readable medium, that can be executed by a computer to restrict access to an instance of a computer program.

7. The system of claim 1, the item of interest comprising a computer system.

8. The system of claim 1, the gesture input interface comprising at least one camera operating in one of the visible and infrared spectra that track the movement of a user within a designated region.

9. The system of claim 1, the gesture input interface comprising a touch pad that is configured to track the position of an input device in contact with the touch pad.

10. The system of claim 1, wherein the item of interest is an entryway, and the access restriction mechanism is an electronic lock.

11. A method for training a gesture based security system, comprising:

generating a plurality of input trajectories from a user representing a characteristic gesture for the user;

normalizing the plurality of input trajectories to produce a set of normalized trajectories having a standard set of dimensions;

reducing each normalized trajectory to a standard set of substantially evenly spaced reference points, each reference point having at least two associated coordinates;

calculating representative statistics for each of the reference spaced points across the plurality of input trajectories to provide a mean for each of the at least two associated coordinates associated with the point, a variance for each of the at least two associated coordinates, and a covariance between at least a first of the at least two coordinates and a second of the at least two coordinates; and

defining respective authentication regions in at least one feature space defined by the associated coordinates of the set of reference points as an authentication range for inputs from the user, the respective authentication regions for the at least one feature space each having at least one associated point from the set of reference points and being defined according to the mean for each of the at least two associated coordinates associated with each of the at least one associated point, a variance for each of the at least two associated coordinates associated with each of the at least one associated point, and a covariance between at least a first of the at least two coordinates and a second of the at least two coordinates associated with each of the at least one associated point.

12. The method of claim 11, wherein receiving an input trajectory representing the characteristic gesture comprises tracking a movement of an input object across a touch screen.

13. The method of claim 11, wherein receiving an input trajectory representing the characteristic gesture comprises tracking the movement of an input object with at least one camera.

14. The method of claim 11, wherein defining a given authentication region of the authentication regions defined in at least one feature space comprises defining a region in feature space in which the Mahalanobis distance between the mean of the region and every point in the region is a threshold value.

15. The method of claim 11, further comprising:

acquiring a gesture from a user as an input trajectory;

normalizing the input trajectory to produce a normalized input trajectory having the standard set of dimensions;

reducing the normalized trajectory to coordinate values for the reference points;

determining if the coordinate values for the reference points fall within the associated authentication regions; and

granting access to an item of interest if the coordinate values for the reference points fall within the associated authentication regions.

16. A security system for authenticating a user, comprising:

an input device configured to a gesture by a user into a trajectory representing a motion of the user over time;

a normalization component configured to resample the trajectory to produce a resampled trajectory having a standard size;

a reference point generator configured to define a standard number of substantially equidistant reference points on the resampled trajectory, each of the reference points having at least two associated coordinate values;

a parameter calculation component configured to calculate representative statistics across a plurality of gestures from a user for the reference points, the representative statistics comprising at least one covariance value between a first coordinate and a second coordinate associated with one of the reference points;

at least one authentication region, each of the at least one authentication region representing at least one of the set of reference points, such that each authentication region is defined within a multidimensional feature space having dimensions corresponding to the at least two coordinates associated with each of the at least one points associated with the authentication region according to the representative statistics for the reference points associated with the authentication; and

a verification component configured to determine if the coordinate values for the set of reference points from a given input fall within the at least one authentication region.

17. The system of claim 16, further comprising an access restriction mechanism configured to restrict access to the item of interest unless the coordinate values for the set of reference points fall within the at least one authentication region.

18. The system of claim 16, wherein the at least one authentication region comprises only one authentication region defined within a multidimensional feature space having dimensions corresponding to the at least two coordinates associated with each point, the verification component being configured to determine if a feature vector representing a given input trajectory falls within the authentication region.

19. The system of claim 16, wherein the at least one authentication region comprises an associated authentication region for each of the set of reference points, each authentication region being defined within a multidimensional feature space having dimensions corresponding to the at least two coordinates associated with each point, the verification component being configured to determine if respective feature vectors representing each reference point from a resampled trajectory falls within its associated authentication region.

20. The system of claim 16, the item of interest comprising one of an entryway or access panel.