Iris liveness detection for mobile devices
An approach for an iris liveness detection is provided. A plurality of image pairs is acquired using one or more image sensors of a mobile device. A particular image pair is selected from the plurality of image pairs, and a hyperspectral image is generated for the particular image pair. Based on, at least in part, the hyperspectral image, a particular feature vector for the eye-iris region depicted in the particular image pair is generated, and one or more trained model feature vectors generated for facial features of a particular user of the device are retrieved. Based on, at least in part, the particular feature vector and the one or more trained model feature vectors, a distance metric is determined and compared with a threshold. If the distance metric exceeds the threshold, then a first message indicating that the plurality of image pairs fails to depict the particular user is generated. It is also determined whether at least one characteristic, of one or more characteristics determined for NIR images, changes from image-to-image by at least a second threshold. If so, then a second message is generated to indicate that the plurality of image pairs depicts the particular user of a mobile device. The second message may also indicate that an authentication of an owner to the mobile device was successful. Otherwise, a third message is generated to indicate that a presentation attack on the mobile device is in progress.
1. A method comprising:
receiving a first image representing at least an eye;
receiving a second image representing at least the eye;
generating, based at least in part on the first image and the second image, a third image representing at least the eye;
generating first data representing a first feature associated with the eye as represented by the third image;
retrieving second data representing a second feature associated with a face of a user;
determining a distance based at least in part on the first feature and the second feature; and
determining whether the eye is associated with the user based at least in part on the distance.
2. The method as recited in claim 1 , wherein:
generating the first data representing the first feature comprises generating a first feature vector based at least in part on the third image;
receiving the second data representing the second feature comprises receiving a second feature vector associated with the face; and
determining the distance comprises determining a distance metric based at least in part on the first feature vector and the second feature vector.
3. The method as recited in claim 1 , wherein determining whether the eye is associated with the user comprises:
determining that the distance is within a threshold; and
based at least in part on the distance being within the threshold, determining that the eye is associated with the user.
4. The method as recited in claim 1 , wherein the first image is generated by a first image sensor at a particular time, and wherein the second image is generated by a second image sensor at the particular time.
5. The method as recited in claim 1 , further comprising:
receiving a fourth image representing at least the eye;
determining a first characteristic associated with the eye as represented by the second image;
determining a second characteristic associated with the eye as represented by the fourth image; and
determining a difference between the first characteristic and the second characteristic,
and wherein determining whether the eye is associated with the user is further based at least in part on the difference.
6. The method as recited in claim 5 , wherein:
the first characteristic includes at least one of a first size of a pupil of the eye, a first pixel intensity associated with the eye, a first hippus of the eye, or a first dilation of the eye; and
the second characteristic includes at least one of a second size of the pupil of the eye, a second pixel intensity associated with the eye, a second hippus of the eye, or a second dilation of the eye.
7. The method as recited in claim 1 , wherein the first image and the second image comprise a first image pair, and wherein the method further comprises:
receiving a second image pair, the second image pair comprising at least a fourth image and a fifth image;
determining that the first image pair represents an iris region associated with the eye;
determining that the second image pair does not represent the iris region associated with the eye; and
selecting the first image pair.
8. The method as recited in claim 1 , further comprising:
receiving a fourth image representing the face of the user;
receiving a fifth image representing the face of the user;
generating, based at least in part on the fourth image and the fifth image, a sixth image representing the face of the user;
generating the second data representing the second feature using at least the sixth image; and
storing the second data representing the second feature.
9. A method comprising:
receiving an image representing at least an eye;
generating, based at least in part on the image, a first feature vector associated with the eye;
retrieving a second feature vector, the second feature vector being associated with a face of a user;
determining a distance based at least in part on the first feature vector and the second feature vector; and
determining whether the eye is associated with the user based at least in part on the distance.
10. The method as recited in claim 9 , wherein the image is a first image, and wherein the method further comprises:
receiving a second image representing at least the eye; and
generating a third image using at least the first image and the second image,
and wherein generating the first feature vector comprises generating the first feature vector based at least in part on the third image.
11. The method as recited in claim 10 , wherein the first image is generated by a first image sensor at a particular time, and wherein the second image is generated by a second image sensor at the particular time.
12. The method as recited in claim 9 , wherein determining whether the eye is associated with the user comprises:
determining that the distance is within a threshold; and
based at least in part on the distance being within the threshold, determining that the eye is associated with the user.
13. The method as recited in claim 9 , further comprising:
receiving an additional image representing at least the eye;
determining a first characteristic associated with the eye as represented by the image;
determining a second characteristic associated with the eye as represented by the additional image; and
determining a difference between the first characteristic and the second characteristic,
and wherein determining whether the eye is associated with the user is further based at least in part on the difference.
14. The method as recited in claim 13 , wherein:
the first characteristic includes at least one of a first size of a pupil of the eye, a first pixel intensity associated with the eye, a first hippus of the eye, or a first dilation of the eye; and
the second characteristic includes at least one of a second size of the pupil of the eye, a second pixel intensity associated with the eye, a second hippus of the eye, or a second dilation of the eye.
15. The method as recited in claim 9 , wherein the image is a first image, and wherein the method further comprises:
receiving a second image representing the face of the user;
receiving a third image representing the face of the user;
generating, based at least in part on the second image and the third image, a fourth image representing the face of the user;
generating the second feature vector using at least the fourth image; and
storing the second feature vector.
16. A method comprising:
receiving a first image representing an eye;
receiving a second image representing the eye;
generating a third image that includes a portion of the first image, the portion of the first image representing the eye;
determining, based at least in part on the third image, that the first image represents a pupil of the eye; and
based at least in part first image representing the pupil of the eye, generating an image pair that includes at least the first image and the second image.
17. The method as recited in claim 16 , further comprising:
generating a fourth image by applying a smoothing function to the third image;
generating a fifth image by applying a derivative operation to the fourth image; and
determining, based at least in part on the fifth image, a magnitude representation of an edge gradient associated with the eye,
and wherein determining that the first image represents the pupil of the eye comprises determining, using the magnitude representation, that the first image represents the pupil of the eye.
18. The method as recited in claim 16 , further comprising:
generating, based at least in part on the first image and the second image, a fourth image representing at least the eye;
generating first data representing a first feature associated with the eye as represented by the fourth image;
retrieving second data representing a second feature associated with a face of a user;
determining a distance based at least in part on the first feature and the second feature; and
determining whether the eye is associated with the user based at least in part on the distance.
19. The method as recited in claim 18 , wherein:
generating the first data representing the first feature comprises generating a first feature vector based at least in part on the fourth image;
receiving the second data representing the second feature comprises receiving a second feature vector associated with the face; and
determining the distance comprises determining a distance metric based at least in part on the first feature vector and the second feature vector.
20. The method as recited in claim 18 , further comprising:
receiving a fifth image representing at least the eye;
determining a first characteristic associated with the eye as represented by the first image;
determining a second characteristic associated with the eye as represented by the fifth image; and
determining a difference between the first characteristic and the second characteristic,
and wherein determining whether the eye is associated with the user is further based at least in part on the difference.