Method of eye aliveness testing and device for eye aliveness testing

• US 8,061,842 B2
• Filed: 08/30/2007
• Issued: 11/22/2011
• Est. Priority Date: 09/07/2006
• Status: Active Grant

• Current Assignee: Naukowa I Akademicka SIEC Komputerowa NASK UL Wawozowa L8
• Original Assignee: Naukowa I Akademicka SIEC Komputerowa NASK UL Wawozowa L8
• Inventors: Chochowski, Marcin, Czajka, Adam, Pacut, Andrzej
• Primary Examiner(s): Dang; Hung
• Application Number:
  US12/440,346
• Publication Number:
  US 20100171925A1
• Time in Patent Office: 1,545 Days
• Field of Search: 351/205, 351/206, 351/221, 351/208, 351/246, 382/2, 382/6, 354/62, 362/227, 362/231
• US Class Current: 351/246
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris
• CPC Class Codes: G06V 40/45 Detection of the body part being alive

Abstract

In compliance with the method, the measurement of the characteristic dimensions of the hypothetical pupil are taken on the basis of a sequence of images. The eye is stimulated with the light featuring a pre-defined intensity profile. For each image in this sequence, the characteristic dimensions of the hypothetical pupil are calculated by means of image processing methods. For a sequence of images, the system determines the function ƒ which defines the changes in the characteristic dimensions of the hypothetical pupil within the measurement period, and on the basis of the said changes as well as on the selected mathematical model, the aliveness parameters O of the eye are determined by means of estimation methods. The calculated aliveness parameters are compared with the statistical template by way of classification process.

Claims

1. A method of automatically testing whether a human eye is alive, the method comprising the steps:

stimulating the pupil of the eye with light radiation emitted by an illuminator at a pre-defined profile of changing light intensity over a time period;

generating a sequence of images of the pupil with a camera during the time period;

calculating characteristic dimensions of a hypothetical pupil from each image in the sequence;

determining a function ƒ

from the sequence of images, wherein ƒ

defines changes in the characteristic dimensions of the hypothetical pupil during the time period;

estimating values for parameters θ

of a pre-defined mathematical eye aliveness model comprising the parameters θ

on the basis of the function ƒ

of the changes of the hypothetical pupil;

comparing the values of the determined parameters θ

for the function ƒ

with a statistical template of values of the parameters θ

for the selected mathematical eye aliveness model by a classification process; and

determining whether the eye is alive based on the comparison.

2. The method of claim 1, wherein the step of calculating characteristic dimensions is performed by a means for digital image processing.

3. The method of claim 1, wherein the characteristic dimensions for each image comprise the average diameter of the pupil, defined as the average value of horizontal and vertical axes of an ellipse that approximates the pupil in the respective image.

4. The method of claim 3, wherein the function ƒ

represents change of the average diameter of the hypothetical pupil within the time period as a reaction to the light stimulation.

5. The method of claim 1, wherein the pre-defined profile of light intensity is a positive stepwise change of light intensity.

6. The method of claim 1, wherein the step of determining the parameters θ

comprises minimizing the sum of the squares of the difference between elements of the calculated sequence of hypothetical pupil average diameters and corresponding average diameter values of a theoretical sequence for a model based on pre-defined parameters.

7. The method of claim 1, wherein the parameters θ

comprise differentiating channel gain, inertial channel gain, differentiating channel time constants, inertial channel time constant, differentiating channel delay, and inertial channel delay.

8. A device for automatic eye aliveness testing comprising the following modules:

an illuminator with light intensity and light color automatic modification systems;

a camera with automatic exposure adjustment connected with the illuminator for obtaining images of a pupil of an eye;

and a measurement and control system connected with the camera and the illuminator;

wherein the measurement and control system comprises;

a converter that converts an analog signal from the camera into a digital image;

a hypothetical pupil localization module, which receives the digital image from the converter and calculates a localized position of the hypothetical pupil within the digital image,

a characteristic dimension determination module, which calculates a characteristic dimension of the pupil based on the digital image from the converter and the localized position from the hypothetical pupil localization module;

a testing controller that receives the localized position from the hypothetical pupil localization module and produces a test initiation signal and sends the test initiation signal to the converter;

an illumination profile controller connected with the illuminator and that receives the test initiation signal from the testing controller and provides in response thereto an illumination intensity profile control signal to the illuminator;

an eye aliveness parameters determination module, which selects a model of pupil dynamics based on the characteristic dimensions of the pupil received from the hypothetical pupil localization module and the illumination intensity profile control signal from the illumination profile controller and which calculates one or more eye aliveness parameters from the model; and

a classifier that receives the eye aliveness parameters from the eye aliveness parameter determination module and parameters from a statistical template and a classification methodology from a classification methodology module and determines whether the eye is alive.