Non-contact type human iris recognition method for correcting a rotated iris image

• US 20020154794A1
• Filed: 12/07/2001
• Published: 10/24/2002
• Est. Priority Date: 03/06/2001
• Status: Abandoned Application

• Current Assignee: Evermedia Company Limited
• Original Assignee: Evermedia Company Limited
• Inventors: Cho, Seong-Won
• Application Number:
  US10/017,118
• Publication Number:
  US 20020154794A1
• Time in Patent Office: Days
• US Class Current: 382/117
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris

Abstract

The present invention relates to an iris recognition method for correcting a rotated iris image. The iris image is acquired by an image acquisition device using an infrared illuminator. The inner and outer boundaries of the iris are detected by analyzing the differences in pixels using a Canny edge detector and the image for the inputted iris image, so as to allow the boundaries of the iris to be more accurately detected from the eye image of a user. If the iris image acquired has been rotated at an angle with respect to the centerline of the iris, the rotated iris image is corrected for subsequent authentication purposes.

Claims

1. A method for correcting a rotated iris image, the method comprising the steps of:

illuminating at least a portion of iris and pupil of a person'"'"'s eye to detect an eye image;

(a) extracting an iris image from said detected eye image;

(b) detecting an inner and outer boundary of said iris image;

(c) converting said extracted iris image into polar coordinates;

(d) if said iris image is slanted, normalizing the corresponding said polar coordinates of said converted iris image so as to yield a predetermined dimension;

(e) comparing previously obtained iris identification information with said normalized polar coordinates obtained in step (d); and

,

(f) determining whether there is a match in data compared in step (e) to authenticate said person.

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

determining whether said iris image is rotated at an angle with respect to the centerline of said iris image;

if yes, temporarily generating a plurality of arrays of said iris image with respect to an array of said converted polar coordinates;

performing a wavelet transform to generate characteristic vectors corresponding to the plurality of said arrays that are temporarily generated;

comparing said respective characteristic vectors generated with previously registered characteristic vectors to obtain similarities; and

,

accepting a new characteristic vector corresponding to the maximum similarity among said obtained similarities as the correct characteristic vector of said person.

3. The method of claim 1, wherein the outer boundary of said iris image is obtained by comparing the pixel value representing said detected eye image with coordinates (x, y) with the other pixel values surrounding the inner boundary of said iris image to determine the maximum difference indicative of the outer boundary of said iris image.

4. The method of claim 1, wherein a predetermined percentage of regions around said iris image is converted into polar coordinates for said comparing step (e).

5. The method of claim 1, wherein an infrared light is used for illumination.

6. The method of claim 1, wherein in a Canny edge detector is used for detecting the inner boundary of said iris image.

7. A method for correcting a rotated iris image, the method comprising the steps of:

(a) capturing a plurality of iris images from a person'"'"'s eye;

(b) detecting an inner and outer boundary of said iris images;

(c) converting the predetermined amount of said captured iris images into polar coordinates;

(d) determining whether one of said iris images is slanted;

(e) if yes, temporarily generating a plurality of arrays of said iris image with respect to an array of said converted polar coordinates;

(f) performing a wavelet transform to generate characteristic vectors corresponding to the plurality of said arrays that are temporarily generated; and

,

(g) comparing said respective characteristic vectors generated with previously registered characteristic vectors to obtain similarities in order to authenticate said person.

8. The method of claim 7, wherein said step (c) comprises the steps of:

normalizing the corresponding said polar coordinates of said converted iris images so as to yield a predetermined dimension;

comparing previously obtained iris identification information with said normalized polar coordinates obtained in said step (c);

determining whether there is a match in data compared in step (e); and

,

accepting a new characteristic vector corresponding to the maximum similarity among said obtained similarities as the correct characteristic vector of said person.

9. The method of claim 7, wherein the outer boundary of said iris image is obtained by comparing the pixel value representing said detected eye image with coordinates (x, y) with the other pixel values surrounding the inner boundary of said iris image to determine the maximum difference indicative of the outer boundary of said iris image.

10. The method of claim 7, wherein in a Canny edge detector is used for detecting the inner boundary of said iris image.