Methods and systems for multiple factor authentication using gaze tracking and iris scanning

• US 7,986,816 B1
• Filed: 09/27/2007
• Issued: 07/26/2011
• Est. Priority Date: 09/27/2006
• Status: Active Grant

• Current Assignee: University of Alaska
• Original Assignee: University of Alaska
• Inventors: Hoanca, Bogdan, Mock, Kenrick
• Primary Examiner(s): Tucker; Wesley
• Assistant Examiner(s): Woldemariam; Aklilu k
• Application Number:
  US11/862,914
• Time in Patent Office: 1,398 Days
• Field of Search: 382/103, 382/115, 382/117, 382/221, 351/209
• US Class Current: 382/115
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities

Abstract

Methods and systems for multiple factor authentication combining eye tracking hardware with iris scanning. The resulting multiple factor authentication is a highly secure and highly accurate authentication procedure. Iris scanning provides excellent identification and eye tracking provides the information that the iris is live and provides identification capabilities based on the eye movement itself while enabling gaze-based password entry.

Claims

1. A method for multiple factor authentication of a user comprising:

a. performing, by a computer, iris recognition for user authentication;

b. performing, by the computer, biometric eye tracking for user authentication;

c. performing, by the computer, password entry via eye tracking for user authentication, wherein performing password entry via eye tracking for user authentication comprises;

determining a plurality of estimated locations where an eye of the user is focused in a series,

receiving data associated with the plurality of estimated locations,

comparing the received data with a reference password,

implementing error correction, wherein the implementing error correction comprises,

determining whether the plurality of estimated locations follows a pattern that comprises k<

S occurrences of event A and S−

k occurrences of event N, wherein k is the number of locations correctly identified by the user, S is the total number of the plurality of estimated locations, A is an event wherein the estimated location is at a correct location, and N is an event wherein the estimated location is at a neighbor location,

determining an authentication probability given by

2. The method of claim 1, wherein performing iris recognition comprises:

capturing an image of an iris of the user;

comparing the image with a reference image; and

successfully authenticating the user if the image matches the reference image.

3. The method of claim 1, further comprising presenting an image to the user, wherein the user is expected to focus on a plurality of specific points in the image in a specific order.

4. The method of claim 1, further comprising presenting a set of tiled images to a user wherein the user is expected to focus on specific tiles in a specific order.

5. The method of claim 1, further comprising presenting an image of a keyboard to a user wherein the user is expected to focus on specific characters of the keyboard in a specific order.

6. The method of claim 1, wherein the probability that the occurrence of the pattern is due to a random try is determined by p_rand(k)=C_S^kp^k(1−

p)^S-k, where

7. The method of claim 1, comprising:

determining, by the computer, a probability over successive authentication sessions that a given pattern of events A and N is due to a user encountering random errors, wherein A is an event wherein the estimated location is at a correct location, and N is an event wherein the estimated location is at a neighbor location; and

successfully authenticating, by the computer, the user if the probability of the pattern being caused by errors is above a predetermined threshold.

8. The method of claim 7, wherein the predetermined threshold is p_rand(k)=C_S^kp^k(1−

p)^S-k, where

9. The method of claim 1, comprising:

determining, by the computer, whether the plurality of estimated locations differs from a plurality of correct locations by a constant translation vector; and

successfully authenticating, by the computer, the user if the plurality of estimated locations differs from the plurality of correct locations by the constant translation vector.

10. The method of claim 1, wherein performing, by the computer, biometric eye tracking comprises extracting user defining information from statistical behavior of the user'"'"'s eye movement.

11. The method of claim 10, further comprising presenting, by the computer, an image that changes in brightness and provides motion for the user to track so as to evoke a response in an eye of the user relating to velocity and pupil diameter.

12. The method of claim 11, further comprising comparing, by the computer, the velocity and pupil diameter of the user to a velocity and a pupil diameter recorded by a known user.

13. A system for multiple factor authentication of a user comprising:

a display device configured for displaying an image to the user;

an eye imaging device configured for taking an image of an iris of the user and for tracking movement of an eye of the user;

a memory configured for storing a reference image, a reference password, a velocity and a pupil diameter of a known user; and

a processor coupled to the display device, the eye imaging device and the memory, the processor configured for

a. performing iris recognition for user authentication,

b. performing biometric eye tracking for user authentication;

c. performing password entry via eye tracking for user authentication, wherein performing password entry via eye tracking for user authentication comprises;

determining a plurality of estimated locations where an eye of the user is focused in a series,

receiving data associated with the plurality of estimated locations,

comparing the received data with a reference password,

implementing error correction, wherein the implementing error correction comprises,

determining whether the plurality of estimated locations follows a pattern that comprises k<

S occurrences of event A and S−

k occurrences of event N, wherein k is the number of locations correctly identified by the user, S is the total number of the plurality of estimated locations, A is an event wherein the estimated location is at a correct location, and N is an event wherein the estimated location is at a neighbor location,

determining an authentication probability given by

14. The system of claim 13, wherein the eye imaging device comprises a digital camera coupled with a near infrared light source.

15. The system of claim 13, wherein successful authentication of a-c comprises

matching an image of an iris of the user if the image matches the reference image;

matching a password entered via eye tracking by the user to the reference password; and

matching a velocity and a diameter of eye movement of the user to the velocity and the pupil diameter of the known user.

16. A non-transitory computer readable storage medium for multiple factor authentication of a user comprising computer executable instruction embodied thereon comprising:

a. performing iris recognition for user authentication,

b. performing biometric eye tracking for user authentication;

c. performing password entry via eye tracking for user authentication, wherein performing password entry via eye tracking for user authentication comprises;

determining a plurality of estimated locations where an eye of the user is focused in a series,

receiving data associated with the plurality of estimated locations,

comparing the received data with a reference password,

implementing error correction, wherein the implementing error correction comprises,

determining whether the plurality of estimated locations follows a pattern that comprises k<

S occurrences of event A and S−

k occurrences of event N, wherein k is the number of locations correctly identified by the user, S is the total number of the plurality of estimated locations, A is an event wherein the estimated location is at a correct location, and N is an event wherein the estimated location is at a neighbor location,

determining an authentication probability given by

17. The non-transitory computer readable storage medium of claim 16, wherein successful authentication of a-c comprises

matching an image of an iris of the user if the image matches the reference image;

matching a password entered via eye tracking by the user to the reference password; and

matching a velocity and a diameter of eye movement of the user to the velocity and the pupil diameter of the known user.

18. A method for multiple factor authentication of a user comprising:

a. performing, by a computer, iris recognition for user authentication;

b. performing, by the computer, biometric eye tracking for user authentication;

c. performing, by the computer, password entry via eye tracking for user authentication, wherein performing password entry via eye tracking for user authentication comprises,

determining a plurality of estimated locations where an eye of the user is focused in a series,

receiving data associated with the plurality of estimated locations,

comparing the received data with a reference password,

implementing error correction, wherein the implementing error correction comprises,

determining a probability over successive authentication sessions that a given pattern of events A and N is due to a user encountering random errors, wherein A is an event wherein the estimated location is at a correct location, and N is an event wherein the estimated location is at a neighbor location, and

successfully authenticating the user if the probability of the pattern being caused by errors is above a predetermined threshold wherein the predetermined threshold is p_rand(k)=C_S^kp^k(1−

p)^S-k, where

19. The method of claim 18, wherein the probability that the occurrence of the pattern is due to a random try is determined by p_rand(k)=C_S^kp^k(1−

p)^S-k, where

20. The method of claim 18, comprising:

determining a probability over successive authentication sessions that a given pattern of events A and N is due to a user encountering random errors, wherein A is an event wherein the estimated location is at a correct location, and N is an event wherein the estimated location is at a neighbor location; and

successfully authenticating the user if the probability of the pattern being caused by errors is above a predetermined threshold.

21. The method of claim 20, wherein the predetermined threshold is p_rand(k)=C_S^kp^k(1−

p)^S-k, where