Rotation compensated iris comparison

• US 20070201728A1
• Filed: 02/27/2006
• Published: 08/30/2007
• Est. Priority Date: 02/27/2006
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: Donald Monro
• Inventors: Monro, Donald
• Granted Patent: US 7,844,084 B2
• Time in Patent Office: Days
• US Class Current: 382/117
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris

Abstract

A method of comparing a test iris image with a candidate iris image comprises generating test and candidate lists representative of measured image values along respective curves around the pupils. Treating these list as periodic, a relative displacement is calculated at which one list best matches the other list. This determines the amount of rotation required for best match of the iris images. The method may be used either alone or in combination with other identity verification systems.

Claims

1. A method of comparing a test iris image with a candidate iris image, comprising the steps of:

(a) generating a test list and a candidate list representative of measured image values along respective curves around respective pupils; and

,

(b) treating the lists as periodic, calculating a relative displacement at which one list best matches the other list, said displacement being representative of a relative rotation at which the test image best matches the iris image.

2. A method as claimed in claim 1 further including the step of sliding the test list with respect to the candidate list.

3. A method as claimed in claim 1 further including the step of calculating a cross-correlation between the test and candidate lists.

4. A method as claimed in claim 1 further including the steps of calculating digital fourier transforms of the test and candidate lists, multiplying the respective transforms or a conjugate thereof, and calculating an inverse digital fourier transform to generate the cross-correlation.

5. A method as claimed in claim 4 further including the step of pre-calculating and storing digital fourier transforms for a plurality of candidate iris images.

6. A method as claimed in claim 1 in which the test and candidate lists are representative of averaged image characteristics over a local region which follows the curves.

7. A method as claimed in claim 1 in which the test and candidate lists represent averaged image values over a plurality of respective curves.

8. A method as claimed in claim 1 in which the test and candidate lists are generated by measuring image values from respective lines within re-mapped rectangular images of the respective irises.

9. A method as claimed in claim 1 further including the step of assessing a quality of match at the said displacement.

10. A method as claimed in claim 9 further including the step of determining whether the test and candidate images do or do not belong to the same individual according to the quality of match.

11. A method as claimed in claim 9 further including the step of using the quality of match as well as additional biometric data to determine whether the test and candidate images do or do not belong to the same individual.

12. A method as claimed in claim 9 further including the steps of comparing the test image with a plurality of candidate images, and determining one or more matches according to the respective qualities of match.

13. A method as claimed in claim 9 further including the steps of comparing the test image with a plurality of candidate images, determining one or more potential matches according to the respective qualities of match, and passing the potential matches to an identity matching system to determine which, if any, of the potential matches can be accepted.