SYSTEMS AND METHODS FOR PERFORMING IRIS IDENTIFICATION AND VERIFICATION USING MOBILE DEVICES

• US 20160117544A1
• Filed: 10/20/2015
• Published: 04/28/2016
• Est. Priority Date: 10/22/2014
• Status: Active Grant

• Current Assignee: Veridium IP Limited (Harvard Investments LP)
• Original Assignee: Hoyos Labs IP, Ltd. (Harvard Investments LP)
• Inventors: Hoyos, Hector, Mather, Jonathan Francis, Xue, Yiqun
• Granted Patent: US 9,767,358 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01B 11/22 for measuring depth
• CPC Class Codes: G06T 2207/30041 Eye; Retina; Ophthalmic
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris
• CPC Class Codes: G06V 40/19 Sensors therefor
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/197 Matching; Classification
• CPC Class Codes: G06V 40/45 Detection of the body part being alive
• CPC Class Codes: H04N 23/11 for generating image signals from visible and infrared light wavelengths
• CPC Class Codes: H04N 23/21 from near infrared [NIR] radiation
• CPC Class Codes: H04N 23/611 where the recognised objects include parts of the human body
• CPC Class Codes: H04N 5/33 Transforming infrared radiation cameras or camera modules for generating image signals from infrared radiation H04N23/20; circuitry of SSIS for transforming infrared radiation into image signals H04N25/20

Abstract

Technologies are presented herein in support of a system and method for performing iris identification and verification using mobile devices. Embodiments of the present invention concern a method for authenticating a user with a mobile device that includes steps for capturing imagery of a user'"'"'s face, analyzing the imagery to identify the portion of the image depicting the user'"'"'s eyes and segmenting the eye image to isolate the iris image data and encoding the iris image data in iris template. The present disclosure further describes authenticating the user based on the encoded iris template and additional techniques for preventing erroneous authentication caused by spoofing. In some examples, the anti-spoofing techniques may include capturing one or more images of a user'"'"'s eyes and analyzing the captured images for indications of liveness based on reflection characteristics depicted in the imagery.

Claims

1. A method for performing iris recognition of a user by a mobile computing device based on imagery of the user captured using the mobile computing device, comprising:

capturing, by the mobile device having a camera, a light emitter, a non-transitory storage medium, instructions stored on the storage medium, and a processor configured by executing the instructions, a first image depicting a facial region of a user;

processing, the first image among the images using an eye finding algorithm to identify an eye of the user depicted in the first image;

illuminating, with the processor using a light source, the user;

capturing, with the processor using the camera during the illuminating step, a second image of the subject;

analyze one or more of the first and second images to identify a location of the eye depicted in the one or more of the first and second images;

isolating, with the processor based on the identified location, an eye image within the second image, wherein the eye image is a portion of the second image that depicts the eye;

segmenting, with the processor, an iris image from the eye image, wherein the iris image is a portion of the eye image that includes iris image data corresponding to an iris;

encoding, with the processor from the segmented iris image, the iris image data as an iris template;

authenticating the user, with the processor, by matching the encoded iris template to a previously stored encoded iris template for a match; and

providing, with the processor, a result of the authentication.

2. The method of claim 1, further comprising:

analyzing one or more of the first and second images to determine that the position of the identified eye is stable.

3. The method of claim 1, wherein the second image is captured at a relatively higher resolution than the first image.

4. The method of claim 1, wherein segmenting the portion of the eye image comprises:

identifying an inner circular boundary between a pupil and the iris and an outer boundary of the iris using a circular gradient algorithm.

5. The method of claim 4, wherein identifying the boundary between the pupil and the iris comprises:

analyzing relative intensities of pixels of the eye image to identify a generally circular inner boundary between the pupil and the iris and a generally circular outer boundary of the iris.

6. The method of claim 1, wherein segmenting the portion of the eye image comprises:

testing a template of a pupil against a plurality of segments of the eye image;

for each test, computing a matching response value corresponding to a closeness of the match; and

identifying the pupil based on the matching response values.

8. The method of claim 1, wherein segmenting the portion of the eye image further comprises:

detecting non-iris features depicted in the eye image, wherein the non-iris features include one or more of;

a specular reflection, an eye lash, and an eyelid; and

modifying, with the processor, one or more of the eye image and the encoded iris template to remove the detected eyelid image data.

9. The method of claim 8, wherein the non-iris features are detected by analyzing relative intensities of pixels of the eye image.

10. The method of claim 8, wherein detecting the eyelid comprises:

analyzing, with the processor, one or more of the eye image and the encoded iris template to detect image data corresponding to an eyelid; and

11. The method of claim 8, wherein the modifying step comprises:

generating, based on the detected non-iris features a mask that obscures image data corresponding to the non-iris features; and

applying the mask to the one or more of the eye image and the encoded iris template.

12. The method of claim 1, further comprising:

normalizing the image data prior to encoding the iris template, wherein normalization includes re-mapping the circular iris image data into an iris strip according to one or more of a pseudo-polar coordinate system or a rectangular coordinate system, and wherein normalization includes scaling the encoded iris template according to a pre-defined size.

13. The method of claim 1, wherein encoding the iris image data as the iris template comprises:

applying, with the processor to the iris image, a Gabor filter to identify iris features; and

generating, with the processor in the memory, the iris template according to the identified iris features.

14. The method of claim 1, wherein the camera is configured to sense one or more of IR light, NIR light, and visible light, and wherein the light emitter is configured to emit one or more of IR light, NIR light and visible light.

15. The method of claim 1, wherein segmenting step is performed for each of a plurality of light channels.

16. The method of claim 1, further comprising:

analyzing the first image to determine an ambient light level;

defining, based on the ambient light level and a output level of the light emitter, one or more settings for the camera, wherein the second image is captured by the camera according to the defined settings, wherein the defined one or more settings include a duration of the flash.

17. The method of claim 1, further comprising:

prompting, with the processor using a display of the mobile device, the user to move the mobile device during the capturing step;

recording, with the processor using a sensor on-board the mobile device, movement data during the capturing step, wherein a plurality of images are captured during the capturing step;

analyzing, with the processor the plurality of images to identify respective positions of ambient light reflections in one or more of the plurality of images; and

comparing the respective positions to the recorded movement data; and

determining whether the respective positions are consistent with the movement of the mobile device.

18. The method of claim 1, wherein the step of providing the result of the authentication includes one or more of:

unlocking, with the processor, one or more applications of the mobile device; and

facilitating, with the processor, access to an access controlled environment.

7. The method of claim, wherein segmenting the portion of the eye image comprises:

testing known samples of iris textures against a plurality of regions of the eye image; and

identifying, based on the testing, regions of the eye image that match the iris texture.

19. A computer implemented method for authenticating a user using a mobile device based on imagery of one or more portions of facial region of the user, wherein the imagery is captured using a camera of the mobile device, the method comprising:

capturing, by a mobile device having a camera, a light emitter, a storage medium, instructions stored on the storage medium, and a processor configured by executing the instructions, a plurality of images depicting at least one facial region of the user, wherein at least a first image is captured by the camera while the light emitter is activated and illuminating the user, and wherein at least a second image is captured while the light emitter is deactivated and is not illuminating the user;

detecting, by the processor in at least the first and the second images from among the plurality of images, an eye of the user depicted in the respective images;

aligning, by the processor, at least the first image and the second image;

subtracting, by the processor, at least a portion of the first image from a corresponding portion of the second image;

detecting, by the processor based on a result of the image subtraction, a reflection characteristic;

determining, by the processor, that the imagery depicts a live user by comparing the detected reflection characteristic to one or more prescribed reflection characteristics.

20. The method of claim 19, wherein the step of aligning is performed using an algorithm that matches the second image to a template.

21. The method of claim 19, wherein the step of aligning at least the first image is performed using an algorithm that matches an image pyramid of the second image to a template of the user'"'"'s eye that was previously recorded on the storage medium.

22. The method of claim 19, wherein the portion of the first image and corresponding portion of the second image each depict at least an iris of the user, and further comprising:

cropping, by the processor, at least the first and second images to isolate the portions of the respective images that depict the iris of the user.

23. The method of claim 19, wherein the step of detecting a reflection characteristic further comprises:

analyzing, by the processor, the result to identify a reflection peak;

calculating, by the processor, a magnitude of the reflection peak; and

wherein liveness is determined as a function of the calculated magnitude of the reflection peak.

24. The method of claim 23, wherein the step of determining liveness further comprises:

calculating, by the processor, a liveness score as a function of the magnitude of the reflection peak;

determining, by the processor, that the calculated liveness score exceeds a prescribed threshold.

25. The method of claim 19, wherein the step of detecting a reflection characteristic further comprises:

analyzing, by the processor, the result to identify a reflection and the reflection characteristic, wherein the reflection characteristic is one or more of;

a size of a reflection;

a location of the reflection;

a shape of the reflection; and

a magnitude (intensity) of the reflection at one or more portions thereof.

26. The method of claim 25, wherein the step of determining liveness further comprises one or more of:

determining that the size of the reflection matches a prescribed size;

determining that the location of the reflection corresponds to a location of a pupil of the user;

determining that the shape of the reflection corresponds to a prescribed shape; and

determining that the reflection exhibits a reflection peak having at least a prescribed magnitude.

27. The method of claim 19, further comprising:

processing, by the processor, at least the first image and the second image using a filter, wherein the filter comprises a band-pass filter applied by the processor to remove reflections and noise present in respective images.

28. The method of claim 19, wherein the step of detecting the eye in at least the first and the second images comprises:

processing, by the processor, a respective image using a face and or eye finder to identify the location of one or more eyes within the respective image; and

determine, by the processor based on the identified location of the one or more eyes in the at least the first and the second images, that the one or more eyes are positionally stable.

29. The method of claim 19, further comprising:

facilitating, by the processor based on the determination that the imagery depicts a live user, the user'"'"'s access to an access controlled environment.

30. The method of claim 19, wherein the step of aligning at least the first and second image is performed, by the processor, based on the orientation of the mobile device during the capture of the images as determined from an onboard position sensor on the mobile device and in communication with the processor.