Detection of an artificial iris for spoofing an iris recognition system

Patent metadata

Field	Value
Publication number	US-12525064-B2
Application number	US-202318199987-A
Country	US
Kind code	B2
Filing date	May 22, 2023
Priority date	May 22, 2023
Publication date	Jan 13, 2026
Grant date	Jan 13, 2026

Abstract

There is provided a computer implemented method of detecting an attempt to breach security of an iris recognition system by an artificial iris, comprising: analyzing at least a portion of a limbal ring depicted in an image of an iris of an individual captured by an imaging sensor at a wavelength range within at least one of near infrared (NIR) and short wave infrared (SWIR), and detecting likelihood of an artificial iris worn by the individual according to the analysis of the at least the portion of the limbal ring.

First claim

What is claimed is:

1 . A computer implemented method of detecting an attempt to breach security of an iris recognition system by an artificial iris, comprising:
analyzing at least a portion of a limbal ring boundary depicted in an image of an iris of an individual captured by an imaging sensor at a wavelength range within at least one of near infrared (NIR) and short wave infrared (SWIR), wherein the limbal ring boundary comprises a transition region between the iris and sclera; computing a sharpness metric that quantifies geometric sharpness of the limbal ring boundary by measuring edge definition characteristics of the transition region between the iris and sclera; and detecting likelihood of an artificial iris worn by the individual according to the analysis of the at least the portion of the limbal ring by comparing the computed sharpness metric to a predetermined threshold value.

2 . The computer implemented method of claim 1 , wherein the wavelength range within at least one of NIR and SWIR is between about 850 nanometers and about 1400 nanometers.

3 . The computer implemented method of claim 1 , further comprising activating an illumination source for illuminating the iris of the individual with an illumination at the wavelength range.

4 . The computer implemented method of claim 1 , wherein when the sharpness metric is above said predetermined threshold the artificial iris is detected as worn by the individual.

5 . The computer implemented method of claim 4 , wherein the image likely excludes the artificial iris when the sharpness metric is below the threshold, for indicating diffusion of the boundary.

6 . The computer implemented method of claim 1 , wherein analyzing the at least the portion of the limbal ring boundary comprises segmenting the portion of the limbal ring, and feeding the segmented at least the portion of the limbal ring into a machine learning model, wherein the detecting the likelihood of the artificial iris is obtained as an outcome of the machine learning model.

7 . The computer implemented method of claim 6 , further comprising training the machine learning model by:
for each sample image of a plurality of sample images of a plurality of sample individuals, each sample image depicting an iris of a sample subject captured by the imaging sensor at the wavelength range within at least one of NIR and SWIR, wherein a first set of sample images depict the sample subjects wearing an artificial iris and a second set of sample images depict the sample subjects that are not wearing the artificial iris:  segmenting at least a portion of the limbal ring, and
creating a training record including the segmented at least the portion of the limbal ring, and a ground truth indicating whether the sample subject is wearing an artificial iris or is not wearing the artificial iris;
creating a multi-record training dataset including a plurality of the training records for the plurality of sample images; and
training the machine learning model on the training dataset.

8 . The computer implemented method of claim 1 , wherein analyzing the at least the portion of the limbal ring boundary comprises:
wherein the image comprises a first image; computing a first sharpness metric indicating sharpness of a boundary of the limbal ring and sclera and/or iris for the first image captured by imaging sensor at the wavelength range within at least one of NIR and short wave infrared SWIR; computing a second sharpness metric indicating sharpness of a boundary of the limbal ring and sclera and/or iris for a second image captured by a second imaging sensor at a second wavelength range within a visible light spectrum; computing a difference between the first sharpness metric and the second sharpness metric; and detecting likelihood of the first image depicting the artificial iris when the difference is above a threshold indicating a non-significant difference between sharpness of the boundary depicted in the first image and sharpness of the boundary depicted in the second image.

9 . The computer implemented method of claim 8 , further comprising detecting likelihood of the first image excluding the artificial iris when the difference is below a threshold indicating a significant difference between diffusion of the boundary depicted in the first image and sharpness of the boundary depicted in the second image.

10 . The computer implemented method of claim 1 , wherein analyzing the at least the portion of the limbal ring boundary comprises detecting a boundary of the artificial iris worn as a contact lens.

11 . The computer implemented method of claim 10 , further comprising computing a sharpness metric indicating sharpness of a boundary of the contact lens, and detecting likelihood of the image depicting the artificial iris when the sharpness metric is above a threshold indicating presence of the contact lens.

12 . The computer implemented method of claim 11 , wherein the image excludes the artificial iris when the sharpness metric is below the threshold, for indicating lack of presence of the contact lens.

13 . The computer implemented method of claim 10 , wherein analyzing the at least the portion of the limbal ring boundary comprises:
wherein the image comprises a first image; computing a first visibility metric indicating visibility of the boundary of the contact lens for the first image captured by imaging sensor at the wavelength range within at least one of NIR and short wave infrared SWIR; computing a second visibility metric indicating visibility of the boundary of the contact lens for a second image captured by a second imaging sensor at a second wavelength range within a visible light spectrum; computing a difference between the first visibility metric and the second visibility metric; and detecting likelihood of the first image depicting the artificial iris when the difference is above a threshold indicating a significant difference between visibility of the boundary depicted in the first image and visibility of the boundary depicted in the second image.

14 . The computer implemented method of claim 13 , further comprising detecting likelihood of the first image excluding the artificial iris when the difference is below a threshold indicating a non-significant difference between visibility of the boundary depicted in the first image and visibility of the boundary depicted in the second image.

15 . The computer implemented method of claim 1 , further comprising, in response to detecting likelihood of the artificial iris, at least one of: generating an alert for presentation on a display, generating instructions for alerting authorities, triggering a second security test, and generating instructions for preventing access to a secure site by the individual.

16 . A system for detecting an attempt to breach security of an iris recognition system by an artificial iris, comprising:
at least one processor executing a code for:
analyzing at least a portion of a limbal ring boundary depicted in an image of an iris of an individual captured by an imaging sensor at a wavelength range within at least one of near infrared (NIR) and short wave infrared (SWIR), wherein the limbal ring boundary comprises a transition region between the iris and sclera;
computing a sharpness metric that quantifies geometric sharpness of the limbal ring boundary by measuring edge definition characteristics of the transition region between the iris and sclera; and
detecting likelihood of an artificial iris worn by the individual according to the analysis of the at least the portion of th

Assignees

• Nec Corp America

Inventors

• Lev Tsvi
• Hoch Yaacov
• Har-Gil Hagai

Classifications

• G06V10/70

  using pattern recognition or machine learning (optical pattern recognition or electronic computations therefor G06V10/88) · CPC title

• G06T2207/10048

  Infrared image · CPC title

• G06T7/10

  Segmentation; Edge detection (motion-based segmentation G06T7/215) · CPC title

• G06V10/141

  Control of illumination · CPC title

• G06T2207/20081

  Training; Learning · CPC title