Texture features for biometric authentication
In general, one aspect of the subject matter described can be embodied in methods that include obtaining one or more image regions from a first image of an eye. Each of the image regions may include a view of a respective portion of the white of the eye. The method may further include applying several distinct filters to each of the image regions to generate a plurality of respective descriptors for the region. The several distinct filters may include convolutional filters that are each configured to describe one or more aspects of an eye vasculature and in combination describe a visible eye vasculature in a feature space. A match score may be determined based on the generated descriptors and based on one or more descriptors associated with a second image of eye vasculature.
1. A computer-implemented method comprising:
obtaining one or more image regions from a first image of an eye, wherein each of the image regions includes a view of a respective portion of vasculature of the eye external to a corneal limbus boundary of the eye;
applying a convolutional filter to each of the image regions at a plurality of different orientations to generate a plurality of respective filtered images for the image region;
generating a respective descriptor for each of the image regions based on, at least, phases of image data elements in the respective filtered images for the image region wherein generating comprises thresholding image data elements of the respective filter images for the image region based on phase values of the image data elements to create respective thresholded images for the image region by setting image data elements having phase values above a first threshold or below a second threshold to a first value, and setting image data elements having phase values in the range of the first threshold to the second threshold to a different second value; and
determining a score based on the generated descriptors and based on one or more descriptors associated with a second image of eye vasculature.
2. The method of claim 1 wherein the first threshold is 0.25 radians and the second threshold is −0.25 radians.
3. The method of claim 1 wherein the first value is one and the second value is zero.
4. The method of claim 1 wherein generating the respective descriptor for each of the image regions comprises:
generating the respective descriptor based on, at least, a mean, median, or magnitude value of image data elements in one or more regions of each of the respective thresholded images for the image region.
5. The method of claim 1 wherein each orientation is 0°, 30°, 60°, 90°, 120°, or 150°.
6. The method of claim 1 wherein the convolutional filter is configured to describe one or more aspects of a visible eye vasculature in a feature space.
7. The method of claim 1 wherein the convolutional filter is a Gabor filter or based on a Gabor filter.
8. The method of claim 1 wherein determining a score further comprises:
creating a respective vector for each of the image regions based on, at least, the respective descriptor for the image region;
for one or more of the image regions of the first image, comparing the respective vector for the image region to a vector derived from descriptors for a corresponding image region of the second image to generate a respective similarity score; and
determining the score based on, at least, the generated similarity scores.
9. The method of claim 1 , further comprising determining whether the score exceeds a value, wherein the value is based on, at least, a robust threshold in a neighborhood of a three-dimensional receiver operating curve generated from a sensitivity analysis.
10. A system comprising:
data processing apparatus programmed to perform operations comprising:
obtaining one or more image regions from a first image of an eye, wherein each of the image regions includes a view of a respective portion of vasculature of the eye external to a corneal limbus boundary of the eye;
applying a convolutional filter to each of the image regions at a plurality of different orientations to generate a plurality of respective filtered images for the image region;
generating a respective descriptor for each of the image regions based on, at least, phases of image data elements in the respective filtered images for the image region wherein generating comprises thresholding image data elements of the respective filter images for the image region based on phase values of the image data elements to create respective thresholded images for the image region by setting image data elements having phase values above a first threshold or below a second threshold to a first value, and setting image data elements having phase values in the range of the first threshold to the second threshold to a different second value; and
determining a score based on the generated descriptors and based on one or more descriptors associated with a second image of eye vasculature.
11. The system of claim 10 wherein the first threshold is 0.25 radians and the second threshold is −0.25 radians.
12. The system of claim 10 wherein the first value is one and the second value is zero.
13. The system of claim 10 wherein generating the respective descriptor for each of the image regions comprises:
generating the respective descriptor based on, at least, a mean, median, or magnitude value of image data elements in one or more regions of each of the respective thresholded images for the image region.
14. The system of claim 10 wherein each orientation is 0°, 30°, 60°, 90°, 120°, or 150°.
15. The system of claim 10 wherein the convolutional filter is configured to describe one or more aspects of a visible eye vasculature in a feature space.
16. The system of claim 10 wherein the convolutional filter is a Gabor filter or based on a Gabor filter.
17. The system of claim 10 wherein determining a score further comprises:
creating a respective vector for each of the image regions based on, at least, the respective descriptor for the image region;
for one or more of the image regions of the first image, comparing the respective vector for the image region to a vector derived from descriptors for a corresponding image region of the second image to generate a respective similarity score; and
determining the score based on, at least, the generated similarity scores.
18. The system of claim 10 , wherein the operations further comprise determining whether the score exceeds a value, wherein the value is based on, at least, a robust threshold in a neighborhood of a three-dimensional receiver operating curve generated from a sensitivity analysis.
19. A non-transitory computer-readable medium having instructions stored thereon that, when executed by one or more data processing apparatus, cause the data processing apparatus to perform operations comprising:
obtaining one or more image regions from a first image of an eye, wherein each of the image regions includes a view of a respective portion of vasculature of the eye external to a corneal limbus boundary of the eye;
applying a convolutional filter to each of the image regions at a plurality of different orientations to generate a plurality of respective filtered images for the image region;
generating a respective descriptor for each of the image regions based on, at least, phases of image data elements in the respective filtered images for the image region wherein generating comprises thresholding image data elements of the respective filter images for the image region based on phase values of the image data elements to create respective thresholded images for the image region by setting image data elements having phase values above a first threshold or below a second threshold to a first value, and setting image data elements having phase values in the range of the first threshold to the second threshold to a different second value; and
determining a score based on the generated descriptors and based on one or more descriptors associated with a second image of eye vasculature.
20. The medium of claim 1 wherein the first threshold is 0.25 radians and the second threshold is −0.25 radians.
21. The medium of claim 19 wherein the first value is one and the second value is zero.
22. The medium of claim 19 wherein generating the respective descriptor for each of the image regions comprises:
generating the respective descriptor based on, at least, a mean, median, or magnitude value of image data elements in one or more regions of each of the respective thresholded images for the image region.
23. The medium of claim 19 wherein each orientation is 0°, 30°, 60°, 90°, 120°, or 150°.
24. The medium of claim 19 wherein the convolutional filter is configured to describe one or more aspects of a visible eye vasculature in a feature space.