IP Library Granted Patent US 7,336,806
Granted Patent B2
US 7,336,806 · App. 10/897,664 · Granted Feb 26, 2008

Iris-based biometric identification

Assignee: Microsoft Corporation
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,336,806
App. No.
10/897,664
Granted
Feb 26, 2008
Kind
B2
Abstract

A method includes forming a feature vector associated with an iris of an individual using a transform operation including converting an iris image into a log-polar image, scaling one or more samples in the log-polar image according to a Fourier-Mellin transform, and performing a Fourier transform on the scaled samples. The method involves forming an iris digest based on the feature vector, and creating a biometric identifier based on the iris digest. A system includes an iris scanner, an iris analyzer that forms an image of a human iris wherein the presence of eyelashes in the image is minimized, forms a feature vector associated with the iris image using a modified Fourier-Mellin transform, and compresses the feature vector into an iris digest.

Claims (54)

1. A method comprising:

forming a feature vector associated with an iris of an individual using a transform operation including:

converting an image of the iris into a log-polar image,

scaling one or more samples in the log-polar image with a Fourier-Mellin transform scalar exponential,

performing a Fourier transform on the log-polar image;

forming an iris digest based on the feature vector; and

creating a biometric identifier based on the iris digest.

2. A method as recited in claim 1 , wherein the biometric identifier is a document.

3. A method as recited in claim 1 , wherein the biometric identifier is an identity card.

4. A method as recited in claim 1 , wherein the scaling is applied in a radial direction of the iris.

5. A method as recited in claim 1 , wherein the converting comprises:

sampling the iris image logarithmically in a radial direction of the iris; and

sampling the iris image uniformly in an annular direction of the iris.

6. A method as recited in claim 1 , wherein the scaling is applied in a radial direction of the iris, and wherein the Fourier transform is applied along an annular direction of the iris.

7. A method as recited in claim 1 , wherein the transform operation is carried out in a discrete domain.

8. A method as recited in claim 1 , wherein the transform operation further comprises:

creating a rectangular representation of the iris; and

masking regions in the rectangular representation that do not include iris data.

9. A method as recited in claim 1 , wherein forming the iris digest based on the feature vector includes compressing the feature vector.

10. A method as recited in claim 1 further comprising:

detecting a pupil in the iris image;

detecting an iris annulus around the iris in the iris image; and

removing noise from the iris image, the noise being due to elements in the iris image other than the iris.

11. A method as recited in claim 1 wherein the feature vector is generated using a distance metric that characterizes a numerical difference between the iris and a reference iris.

12. A computer-readable medium having stored thereon computer-executable instructions for causing a computer to perform a process comprising:

forming an isolated iris image from an image of an eye of an, individual;

forming an iris digest based on the isolated iris image using a transform operation including:

converting isolated iris image into a log-polar image,

scaling one or more samples in the log-polar image according to a Fourier-Mellin transform,

performing a Fourier transform on the log-polar image; and

creating a biometric identifier based on the iris digest.

13. A computer-readable medium as recited in claim 12 , wherein forming an isolated iris image comprises:

detecting a pupil in the image of the eye;

detecting an iris annulus around an edge of the iris in the image of the eye;

removing noise from the image of the eye around the iris annulus.

14. A computer-readable medium as recited in claim 12 , the process further comprising:

hashing text descriptive of the individual; and

combining the hashed text with the iris digest.

15. A computer-readable medium as recited in claim 12 , the process further comprising signing the biometric identifier with a private-key.

16. A computer-readable medium as recited in claim 12 , the process further comprising encoding the biometric identifier in a barcode.

17. A computer-readable medium as recited in claim 12 , the process further comprising:

encoding the biometric identifier in a barcode; and

printing the barcode on a document.

18. A computer-readable medium as recited in claim 12 , the process further comprising:

hashing text descriptive of the individual;

combining the hashed text with the iris digest;

including the combined hashed text and the iris digest in the biometric identifier;

signing the biometric identifier with a private-key; and

encoding the biometric identifier in a barcode.

19. A computer-readable medium as recited in claim 12 , wherein forming the iris digest further includes creating a feature vector based on the Fourier transform.

20. A computer-readable medium as recited in claim 12 , wherein forming the iris digest further includes creating a feature vector by selecting one or more Fourier transform coefficients generated from the Fourier transform.

21. A computer-readable medium as recited in claim 12 , wherein the scaling is applied in a radial direction of the iris, and wherein the Fourier transform is applied along an annular direction of the iris.

22. A computer-readable medium as recited in claim 12 , wherein the process further comprises forming a face digest from an image of a face of the individual.

23. A computer-readable medium as recited in claim 12 , wherein the process further comprises forming a face digest from an image of a face of the individual, and wherein the creating a biometric identifier comprises creating a biometric identifier based on the iris digest and the face digest.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 9, 2014
From: MICROSOFT CORPORATION
To: MICROSOFT TECHNOLOGY LICENSING, LLC
Reel/Frame 034541/0477 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 22, 2004
From: SCHONBERG, DANIEL; KIROVSKI, DARKO
To: MICROSOFT CORPORATION
Reel/Frame 015619/0815 →
Continuity (2)
Provisional Application 6055555500 · Mar 22, 2004
Related Publication 20050207614A1 · Sep 22, 2005