Binding of selfie face image to iris images for biometric identity enrollment

Abstract

An imaging system associates one or more the iris images with a selfie face image for biometric identity. The imaging system captures a first image of a subject using light in a visible wavelength range and captures a second image of the subject using light in a near infrared wavelength range. The image system retrieves a selfie face image of the subject, and matches a face portion of the first image against a face portion of the selfie face image, wherein the face portion for each at least comprises an iris. In response to the matching, the image system segments one or more iris portions from the second image to generate one or more iris images and binds the first image and the one or more iris images to the selfie face image for biometric identity.

What is claimed is: 
     
         1 . A method comprising:
 capturing, with an imaging system, a three-dimensional (3D) illumination intensity map of a subject's face, wherein the 3D illumination intensity map comprises an illumination intensity at each of a plurality of 3D positions on the subject's face;   determining at least one eye location of at least one eye of a subject based on the 3D illumination intensity map, the location comprising a distance between the imaging system and the subject's one or more eyes;   determining an average pixel value of a portion of the subject's face including the at least one eye;   converting the 3D illumination intensity map into a 2D illumination intensity map including the determined eye locations;   normalizing the 2D illumination intensity map based on the average pixel value; and   capturing, with the imaging system, an iris image based on the normalized 2D illumination intensity map and the determined at least one eye location.   
     
     
         2 . The method of  claim 1 , wherein determining the average pixel value of the face's portion containing the one or more eyes comprises:
 cropping out the face's portion containing the one or more eyes based on the determined eye locations.   
     
     
         3 . The method of  claim 1 , wherein converting the 3D illumination intensity map into the 2D illumination intensity map including the determined eye locations comprises is based in part on a look-up table that stores a relationship between the 3D illumination intensity map and the 2D illumination intensity map for each determined eye location. 
     
     
         4 . The method of  claim 1 , wherein normalizing the 2D illumination intensity map is further based on a difference in quantum efficiency between a 3D imager configured to generate the 3D illumination intensity map and an iris imager configured to capture the iris image. 
     
     
         5 . The method of  claim 1 , wherein normalizing the 2D illumination intensity map is further based on a ratio in exposure between a 3D imager configured to generate the 3D illumination intensity map and an iris imager configured to capture the iris image. 
     
     
         6 . The method of  claim 1 , wherein capturing the iris image based on the normalized 2D illumination intensity map and the determined eye locations such that the captured iris image is not saturated. 
     
     
         7 . The method of  claim 1 , wherein capturing the iris image based on the normalized 2D illumination intensity map and the determined eye locations such that a corneal glint in the captured iris image is not saturated. 
     
     
         8 . The method of  claim 1 , wherein capturing the iris image based on the normalized 2D illumination intensity map and the determined eye locations comprises:
 determining a focus position range based on the determined eye locations; and   adjusting a focus position in the focus position range to capture the iris image.   
     
     
         9 . The method of  claim 1 , further comprising:
 locating the subject's face using a face-finding software; and   capturing a face image of the subject's face based on the locating.   
     
     
         10 . The method of  claim 1 , further comprising:
 capturing, with the imaging system, a second image of the subject using light in a visible wavelength range;   capturing, with the imaging system, a third image of the subject using light in a near infrared wavelength range;   retrieving a selfie face image of the subject, wherein the selfie face image was previously captured using an imaging device that is different from the imaging system;   matching a face portion of the second image against a face portion of the selfie face image;   segmenting one or more iris portions from the third image to generate one or more iris images; and   storing an association between (i) the one or more iris images and the (ii) selfie face image.   
     
     
         11 . A system comprising:
 a three-dimensional (3D) imager configured to generate a 3D illumination intensity map of a subject's face, wherein the 3D illumination intensity map comprises an illumination intensity at each 3D position of the subject's face;   an iris imager configured to capture an iris image of the subject's one or more eyes;   a processor configured to execute computer-executable instructions for performing a method comprising:
 determining eye locations of the subject's one or more eyes based on the 3D illumination intensity map, wherein the locations at least comprise a distance between the 3D imager and the subject's one or more eyes; 
 determining an average pixel value of a face's portion containing at least one of the subject's eyes; 
 converting the 3D illumination intensity map into a 2D illumination intensity map including the determined eye location; 
 normalizing the 2D illumination intensity map based on the average pixel value; and 
 capturing the iris image based on the normalized 2D illumination intensity map and the determined eye locations using the iris imager. 
   
     
     
         12 . The system of  claim 11 , wherein determining the average pixel value of the face's portion containing the one or more eyes comprises:
 cropping out the face's portion containing the one or more eyes based on the determined eye locations.   
     
     
         13 . The system of  claim 11 , wherein converting the 3D illumination intensity map into the 2D illumination intensity map including the determined eye locations comprises is based in part on a look-up table that represents a relationship between the 3D illumination intensity map and the 2D illumination intensity map for each determined eye location. 
     
     
         14 . The system of  claim 11 , wherein normalizing the 2D illumination intensity map is further based on a difference in quantum efficiency between the 3D imager and the iris imager configured to capture the iris image. 
     
     
         15 . The system of  claim 11 , wherein normalizing the 2D illumination intensity map is further based on a ratio in exposure between the 3D and the iris imager. 
     
     
         16 . The system of  claim 11 , wherein capturing the iris image based on the normalized 2D illumination intensity map and the determined eye locations such that the captured iris image is not saturated. 
     
     
         17 . The system of  claim 11 , wherein capturing the iris image based on the normalized 2D illumination intensity map and the determined eye locations such that a corneal glint in the captured iris image is not saturated. 
     
     
         18 . The system of  claim 11 , wherein capturing the iris image based on the normalized 2D illumination intensity map and the determined eye locations comprises:
 determining a focus position range based on the determined eye locations; and   adjusting a focus position in the focus position range to capture the iris image.   
     
     
         19 . The system of  claim 11 , wherein the 3D imager comprises a stereo camera. 
     
     
         20 . The system of  claim 11 , further comprising a face imager configured to capture a face image of the subject's face based on the determined eye locations.