Iris recognition method and apparatus thereof

• US 20080069410A1
• Filed: 02/06/2007
• Published: 03/20/2008
• Est. Priority Date: 09/18/2006
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Gil, Youn Hee, Yoo, Jang Hee, Ko, Jong Gook, Chung, Kyo II
• Granted Patent: US 7,869,626 B2
• Time in Patent Office: Days
• US Class Current: 382/117
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction

Abstract

An iris recognition method is provided. In the iris recognition method, binary image data is obtained by receiving eye image data and filtering the received eye image data using a predetermined threshold value. Then, candidate center search regions are searched for finding a pupil center using profile information of rows and columns of the binarized image. a pupil boundary and a center point are detected by performing a mask operation using 8 pupil boundary mask templates for each of concentric circles formed of pointes in the searched candidate center search regions as candidate pupil centers and different radiuses. An iris boundary region is detected by performing a masking operation using 6 iris boundary mask templates corresponding to 6 locations for concentric circles formed of the pupil center as an origin and different radiuses.

Claims

1. An iris recognition method comprising:

obtaining binary image data by receiving eye image data and filtering the received eye image data using a predetermined threshold value;

searching candidate center search regions for finding a pupil center using profile information of rows and columns of the binarized image;

detecting a pupil boundary and a center point by performing a mask operation using 8 pupil boundary mask templates for each of concentric circles formed of pointes in the searched candidate center search regions as candidate pupil centers and different radiuses; and

detecting an iris boundary region by performing a masking operation using 6 iris boundary mask templates corresponding to 6 locations for concentric circles formed of the pupil center as an origin and different radiuses.

2. The method according to claim 1, wherein in the step of obtaining the binarized image, pixel value distribution in the eye image data are analyzed, pixels having values smaller than a threshold value are transformed to 1, and pixels having values larger than a threshold value are transformed to 0, thereby binarizing the eye image data.

3. The method according to claim 1, wherein the step of detecting the pupil boundary and the center point includes multiplying the eight pupil boundary masks and the search areas in an iris image for circles that are searched by a radius r and a center point (x, y), adding the multiplying results, and setting a radius and an origin of a pupil as a radius r and a center coordinate (x, y) of a circle having the largest value using Equation:

$\underset{\left(r,\left(x,y\right)\right)}{\mathrm{Max}}<br><br>\left[\underset{i=1}{\overset{8}{\sum <br><br>}}<br><br>\mathrm{pupil\_boundary}<br><br>\mathrm{\_mask}<br><br>\left(i\right)\times <br><br>\mathrm{img\_area}<br><br>\left(i\right)\right],$

wherein pupil_boundary_mask(i) denotes the eight pupil boundary masks, and image_area(i) denotes search regions.

4. The method according to anyone of claims 1 and 3, wherein the eight pupil boundary mask templates are separated from a circle at 45°

.

5. The method according to claim 1, wherein the pupil boundary mask template is formed of a 15×

15 formation.

6. The method according to claim 5, wherein in the pupil boundary mask template, bright color pixels have positive coefficient values, and dark color pixels have negative coefficient values.

7. The method according to claim 1, wherein bright color pixels of the upper most pupil boundary mask template and the lower pupil boundary mask template are smaller than those of other pupil boundary mask templates.

8. The method according to claim 1, wherein the six iris boundary mask templates are two contact points of extension of a horizontal line passing a pupil center point and a circle having the pupil center point as a center, and boundary points of circles separated from the contact points at about 22.5°

upwardly and downwardly.

9. The method according to claim 1, wherein the iris boundary mask template is formed of a 15×

15 formation.

10. The method according to claim 9, wherein in the iris boundary mask template, bright color pixel values have positive coefficient values and dark color pixel values have negative coefficient values.

11. The method according to claim 1, wherein the iris boundary is detected by multiplying the six iris boundary masks and the search areas in binary image that is an object of each masking operation for circles searched by a radius r and a center point (x, y), adding the multiplying results, and deciding a radius r of a circle having the largest value as an iris boundary using Equation:

$\underset{\left(r,\left(x,y\right)\right)}{\mathrm{Max}}<br><br>\left[\underset{i=1}{\overset{6}{\sum <br><br>}}<br><br>\mathrm{iris\_boundary}<br><br>\mathrm{\_mask}<br><br>\left(i\right)\times <br><br>\mathrm{img\_area}<br><br>\left(i\right)\right],$

wherein iris_boundary_mask(i) denotes the six iris boundary masks, and image_area(i) denotes the search regions.

12. An iris recognition apparatus comprising:

an image input unit for receiving eye image data obtained by photographing eyes using an external optical device;

a binarization unit for binarizing the input image data through filtering;

a pupil boundary search unit including a template matching module for comparing pupil boundary mask templates corresponding to 8 locations of a pupil boundary region with the iris image data for detecting a pupil boundary and a center point using histogram information from the binarized images; and

an iris boundary search unit including a template matching module for searching an iris boundary using iris boundary mask templates corresponding to six locations in a registered iris boundary region.

13. The iris recognition apparatus according to claim 12, the binarization unit analyzes pixel distribution values in the eye image data, transforms pixels having values smaller than a threshold value to 1, and pixels having values larger than a threshold value to 0.