Multimodal ocular biometric system and methods
First Claim
1. A method for segmenting a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
- defining one or more first search windows in a captured eye image;
identifying a first set of points corresponding to peaks in an image intensity gradient in each of the one or more first search windows;
determining, from the first set of points, segments according to image intensity transitions, each segment having a center point;
grouping sets of the segments into a first set of clusters according to positions of the center points for the segments; and
selecting, from the first set of clusters, a second set of clusters corresponding to a pupil image;
fitting a first model template to the points for each set of segments corresponding to the second set of clusters, the fitted first model template corresponding to an initial pupil outer boundary for the pupil image;
defining one or more second search windows in relation to the initial pupil outer boundary;
identifying a second set of points corresponding to peaks in an image intensity gradient in each of the one or more second search windows; and
fitting a second model template to the second set of points, the fitted second model template corresponding to a refined pupil outer boundary for the pupil image.
7 Assignments
0 Petitions
Accused Products
Abstract
Biometric systems capture and combine biometric information from more than one modality, employing digital processing algorithms to process and evaluate captured images having data for a biometric characteristic. Such digital algorithms may include a pupil segmentation algorithm for determining a pupil image in the captured image, an iris segmentation algorithm for determining an iris image in the captured image, an eyelid/eyelash segmentation algorithm for determining an eyelid/eyelash image in the captured image, and an algorithm for measuring the focus on the iris. Some embodiments employ an auto-capture process which employs such algorithms, in part, to evaluate captured images and obtain the best possible images for biometric identification.
156 Citations
64 Claims
-
1. A method for segmenting a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
-
defining one or more first search windows in a captured eye image; identifying a first set of points corresponding to peaks in an image intensity gradient in each of the one or more first search windows; determining, from the first set of points, segments according to image intensity transitions, each segment having a center point; grouping sets of the segments into a first set of clusters according to positions of the center points for the segments; and selecting, from the first set of clusters, a second set of clusters corresponding to a pupil image; fitting a first model template to the points for each set of segments corresponding to the second set of clusters, the fitted first model template corresponding to an initial pupil outer boundary for the pupil image; defining one or more second search windows in relation to the initial pupil outer boundary; identifying a second set of points corresponding to peaks in an image intensity gradient in each of the one or more second search windows; and fitting a second model template to the second set of points, the fitted second model template corresponding to a refined pupil outer boundary for the pupil image. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14)
-
-
15. A method for segmenting a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
-
receiving image data for a pupil image of a captured eye image, the image data including a pupil outer boundary; defining one or more first search windows in relation to the pupil image; identifying a first set of points corresponding to image intensity transitions to a right side of the pupil image and to a left side of the pupil image, the set of points corresponding to edge points for an iris image; fitting a first model template to the set of points, the fitted first model template corresponding to an initial iris outer boundary for the iris image; defining one or more second search windows in relation to the initial iris outer boundary for the iris image; identifying a set of points corresponding to peaks in an image intensity gradient in each of the one or more second search windows; and fitting a second model template to the fourth set of points, the fitted second model template corresponding to a refined iris outer boundary for the iris image. - View Dependent Claims (16, 17, 18, 19, 20)
-
-
21. A method for segmenting a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
-
receiving image data for an iris image of a captured eye image, the image data including an inner iris boundary and an outer iris boundary defining an iris annular region; determining an image intensity distribution in one or more areas of the iris annular region below the inner iris boundary; determining iris intensity thresholds according to the image intensity distribution; identifying non-iris pixels above and below the inner iris boundary by applying the iris intensity thresholds to the eye image; detecting, from the non-iris pixels, a first set of edge points in search windows to the left of the annular region and to the right of the annular region; reducing the first set of edge points to a second set of edge points corresponding to a boundary of occluded regions in the eye image; dividing the second set of edge points between a first class corresponding to an upper eyelid image in the eye image and a second class corresponding to a lower eyelid in the eye image; fitting a first curve to the points in the first class; fitting a second curve to the points in the second class; determining intersections between the first curve, second curve, the refined iris inner boundary, and the refined iris outer boundary; integrate an area between the first curve, second curve, the refined iris inner boundary, and the refined iris outer boundary, the area corresponding to the area of the occluded regions. - View Dependent Claims (22)
-
-
23. A method for segmenting a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
-
receiving image data for an iris image of a captured eye image, the image data including an inner iris boundary and an outer iris boundary defining an iris annular region; unwrapping the annular iris region into a rectangular iris image; determining training set histograms for regions in the rectangular iris image free of eyelid or eyelash occlusion; selecting a set of test points within the eye image for testing for occlusion by eyelids and eyelashes; defining test regions according to the set of test points; determining a test histogram for each test region; determining a histogram intersection score, each test histogram with the training set histograms; and determining whether each test region corresponds to an eyelid or an eyelash occlusion, comprising the step of comparing the histogram intersection score of each test region to an overlap threshold value. - View Dependent Claims (24, 25, 26, 27, 28, 29, 30)
-
-
31. A method for segmenting a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
-
receiving image data for an iris image of a captured eye image, the image data including an inner iris boundary and an outer iris boundary defining an iris annular region; unwrapping the annular iris region into a rectangular iris image; determining a phase congruency map according to a bank of log-Gabor filters; determining a first weighted texture intensity image corresponding to dark areas in the rectangular iris image; determining a second weighted texture intensity image corresponding to light areas in the rectangular iris image; determining an initial mask image from the first weighted texture intensity image and the second weighted texture intensity image, the initial mask image marking occlusion pixels and iris pixels; and processing outlier pixels in the initial mask to produce a final mask image. - View Dependent Claims (32, 33, 34, 35, 36, 37, 38)
-
-
39. A method for measuring focus of a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
-
receiving image data for an iris image of a captured eye image, the image data including an inner iris boundary and a center of the inner iris boundary; determining a gradient magnitude across the inner iris boundary in a radial direction with respect to the pupil center; determining a histogram of the gradient magnitude; and identifying a value corresponding to a percentile of the gradient magnitude, the value corresponding to a focus measure. - View Dependent Claims (40, 41)
-
-
42. A method for measuring focus of a biometric characteristic from an eye image captured by a biometric device, the method comprising the steps of:
-
determining a measured light reflection size from a captured eye image; and normalizing the measured light reflection size with a best focus reflection size corresponding to a best focus point, the normalized measured light reflection size corresponding with the focus measure.
-
-
43. A method for processing an eye image captured by a biometric device, the method comprising:
-
receiving a captured frame containing an eye image; applying a segmentation test to the eye image, the segmentation test providing segmentation data; and applying, if the eye image passes the segmentation test, an image quality test to the eye image, the image quality test providing image quality data; adding, if the eye image passes the image quality test, an image acquisition result to an image data cache, the image acquisition result including the eye image, the segmentation data, and the image quality data corresponding to the eye image, and the image data cache containing cached data acquisition results. - View Dependent Claims (44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64)
-
Specification