Apparatuses, systems, and methods for confirming identity

Patent metadata

Field	Value
Publication number	US-9922238-B2
Application number	US-201514841581-A
Country	US
Kind code	B2
Filing date	Aug 31, 2015
Priority date	Jun 25, 2015
Publication date	Mar 20, 2018
Grant date	Mar 20, 2018

Abstract

A system, apparatus, and method for confirming the identity of a person for access using an image of that person. The system includes a camera, a data storage device, a subject identification device, and a processor to compare a plurality of current images of the subject to a plurality of stored images of that subject and confirm that the subject live and an approved subject.

First claim

What is claimed is:

1. A live human detection system, comprising:
a camera for capturing a plurality of current images of a human subject;
a data storage device for storing a plurality of stored images of the human subject, each stored image including at least one of features of at least one eye of the human subject and data created from an image of at least one eye of the human subject;
a subject identification device; and
a processor coupled to the data storage device, the camera, and the subject identification device, the processor including instructions which, when executed by the processor, cause the processor to:
receive a subject identification from the subject identification device;
capture a plurality of current images from the camera, those images including at least one initial image, at least one intermediate image, and at least one later image, where at least one of the plurality of current images can be both an initial image and an intermediate image and at least one other of the plurality of current images can be both an intermediate image and a later image;
convert each of the plurality of current images received from the camera to greyscale if the plurality of current images are in color;
detect at least one of the eyes of the subject in each of the plurality of current images;
determine whether an ocular region including one of the detected eyes in the at least one initial image matches a corresponding open eye in at least one of the stored images of the subject;
determine whether an ocular region including one of the detected eyes in at least one intermediate image matches a corresponding closed eye in at least one of the stored images of the subject;
determine whether an ocular region including one of the detected eyes in at least one later image matches the corresponding open eye in at least one of the stored images of the subject; and
conclude that the subject is live and an approved subject if the ocular region including the at least one eye in the at least one initial image matches the corresponding open eye in at least one of the stored images of the subject, the ocular region including the at least one eye in at least one intermediate image matches a corresponding closed eye in at least one of the stored images of the subject, and ocular region including the at least one eye in at least one later image matches the corresponding open eye in at least one of the stored images of the subject.

2. The live human detection system of claim 1 , further comprising:
an input device coupled to the processor; and
a display coupled to the processor; and
wherein the processor further includes instructions which, when executed by the processor, cause the processor to:
display one or more approved subjects on the display;
receive an input from the input device that indicates which displayed approved subject is the human subject presented; and
retrieve the stored image of the selected approved subject with one or more eyes closed and the stored image of the selected approved subject with one or more eyes open.

3. The live human detection system of claim 1 , wherein the processor further includes instructions which, when executed by the processor, cause the processor to:
binarize each of the plurality of current images;
group contiguous pixels in each of the binarized plurality of current image into segments;
identify features of the ocular region including one of the detected eyes in each of the plurality of current images;
remove segments having fewer than a predetermined number of contiguous pixels from the features of the ocular region including one of the detected eyes;
mask each of the plurality of current images to remove segments around the ocular region including one of the detected eyes.

4. The system of claim 3 , wherein the grouping of the contiguous pixels includes top hat segmenting the current image.

5. The live human detection system of claim 1 , further comprising:
a blink-initiation device coupled to the processor; and
wherein the processor further includes instructions which, when executed by the processor, cause the processor to:
receive a blink-initiation signal from the blink initiation device; and
receive the plurality of current images from the camera after receiving the blink-initiation signal.

6. The live human detection system of claim 5 , further comprising:
a speaker coupled to the processor; and
wherein the processor further includes instructions which, when executed by the processor, cause the processor to:
provide an audible countdown through the speaker; and
capture the plurality of images at at least one of during the countdown and after the countdown.

7. The live human detection system of claim 1 , wherein the subject identification device is a touchscreen display for displaying list of approved subjects and on which the subject to be confirmed is to be selected from the list.

8. The live human detection system of claim 7 , wherein the at least one later image is at least one of the last of the plurality of current images captured and at least one image captured consecutively before the last of the plurality of current images captured.

9. The live human detection system of claim 8 , wherein the at least one initial image is at least one of the first of the plurality of current images captured and at least one image captured consecutively after the first of the plurality of current images captured.

10. The live human detection system of claim 1 , wherein the at least one later image includes the last of the plurality of current images captured.

11. The live human detection system of claim 1 , wherein the at least one initial image includes the first of the plurality of current images captured.

12. The live human detection system of claim 1 , wherein the at least one intermediate image is captured after the first of the plurality of current images is captured and before the last of the plurality of current images is captured.

13. The live human detection system of claim 1 , wherein the processor further includes instructions which, when executed by the processor, cause the processor to mask the ocular region surrounding and including the at least one eye after detecting the at least one eye of the subject in each of the plurality of images.

14. The live human detection system of claim 1 , wherein the processor further includes instructions which, when executed by the processor, cause the processor to:
detect both eyes of the subject in each of the plurality of current images;
determine whether both eyes of the subject in an initial image match both corresponding open eyes in a stored image of the subject;
determine whether both eyes in an intermediate image match both corresponding closed eyes in a stored image of the subject;
determine whether both eyes in a later image match both corresponding open eyes in the stored image of the subject;
conclude that the subject is live and an approved subject if both eyes in the initial image match both corresponding open eyes in the stored image of the subject, both eyes in the intermediate image match both corresponding closed eyes in the stored image of the subject, and both eyes in the later image match both corresponding open eyes in the stored image of the subject.

15. The live human detection system of claim 14 , wherein the function is an application.

16. The live human detection system of claim 1 , wherein the processor further includes instructions which, when executed by the processor, cause the processor to access a function of a computing device.

Assignees

• Univ West Virginia

Inventors

• Bourlai Thirimachos
• Wowczuk Zenovy S
• Ferrell Walter A
• Amerman Steven R
• Wilhelm Jay P
• Osia Nnamdi C
• Esposito Ii Patrick R
• Upton Terrence H

Classifications

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title

• G06K9/00221

  Physics · mapped topic

• G06K9/00926

  Physics · mapped topic

• G06K9/00228

  Physics · mapped topic

• G06T2207/20221

  Image fusion; Image merging · CPC title