Quality metrics for biometric authentication

1. A computer-implemented method comprising:

obtaining a first image of an eye, wherein the first image includes a view of a portion of a vasculature of the eye external to a corneal limbus boundary of the eye;

determining a plurality of metrics for the first image, wherein the metrics include a first metric for reflecting an extent of one or more connected structures in the first image that represents a morphology of eye vasculature and a second metric for comparing the extent of eye vasculature detected across different color components in the first image by determining a relationship of a variation of a first color difference signal to a variation of a second color difference signal;

determining a quality score based on, at least, the plurality of metrics for the first image, wherein the quality score is a prediction of a match score that would be determined based on the first image and a second image, assuming the first image and the second image included a view of the same person's vasculature; and

rejecting or accepting the first image based on the quality score.

2. The method of claim 1 , in which determining the first metric comprises:

dilating the vasculature in the first image so that one or more disconnected veins in the vasculature become connected;

thinning the dilated vasculature in the first image; and

determining a portion of the first image that contains the thinned dilated vasculature.

3. The method of claim 1 wherein determining a relationship of a variation of a first color difference signal to a variation of a second color difference signal comprises:

determining the first color difference signal by subtracting a first color component of the first image from a second color component of the first image;

determining the second color difference signal by subtracting a third color component of the first image from the first color component of the first image; and

calculating a ratio of a standard deviation of the first color difference signal to a standard deviation of the second color difference signal.

4. The method of claim 1 , in which the plurality of metrics includes a third metric based on an amount of glare in the first image.

5. The method of claim 1 , in which the plurality of metrics includes a third metric based on a number and types of occlusions in the first image.

6. The method of claim 1 , in which the plurality of metrics includes a third metric based on a gaze angle of an eye depicted in the first image.

7. The method of claim 1 , in which the plurality of metrics includes a third metric based on the segmentation quality of a sclera depicted in the first image.

8. The method of claim 1 , further comprising determining a match score by combining, based in part on the quality score, a plurality of match scores, including at least one match score based on the first image.

9. The method of claim 1 , further comprising:

determining a match score based on, at least, the first image and data from a reference record that reflects a reference image;

accepting a user based in part on the match score;

comparing the quality score to a previous quality score stored in the reference record; and

updating the reference record with data based on the first image when the quality score is better than the previous quality score.

10. The method of claim 1 , further comprising providing feedback based on the quality score to a user.

11. A system, comprising:

a data processing apparatus; and

a memory coupled to the data processing apparatus having instructions stored thereon which, when executed by the data processing apparatus cause the data processing apparatus to perform operations comprising:

obtaining a first image of an eye, wherein the first image includes a view of a portion of a vasculature of the eye external to a corneal limbus boundary of the eye;

determining a plurality of metrics for the first image, wherein the metrics include a first metric for reflecting an extent of one or more connected structures in the first image that represents a morphology of eye vasculature and a second metric for comparing the extent of eye vasculature detected across different color components in the first image by determining a relationship of a variation of a first color difference signal to a variation of a second color difference signal;

determining a quality score based on, at least, the plurality of metrics for the first image, wherein the quality score is a prediction of a match score that would be determined based on the first image and a second image, assuming the first image and the second image included a view of the same person's vasculature; and

rejecting or accepting the first image based on the quality score.

12. The system of claim 11 , in which determining the first metric comprises:

dilating the vasculature in the first image so that one or more disconnected veins in the vasculature become connected;

thinning the dilated vasculature in the first image; and

determining a portion of the first image that contains the thinned dilated vasculature.

13. The system of claim 11 , wherein determining a relationship of a variation of a first color difference signal to a variation of a second color difference signal, in which determining the second metric comprises:

determining the first color difference signal by subtracting a first color component of the first image from a second color component of the first image;

determining the second color difference signal by subtracting a third color component of the first image from the first color component of the first image; and

calculating a ratio of a standard deviation of the first color difference signal to a standard deviation of the second color difference signal.

14. The system of claim 11 , in which the plurality of metrics includes a third metric based on an amount of glare in the first image.

15. The system of claim 11 , in which the plurality of metrics includes a third metric based on a number and types of occlusions in the first image.

16. The system of claim 11 , in which the plurality of metrics includes a third metric based on a gaze angle of an eye depicted in the first image.

17. The system of claim 11 , in which the plurality of metrics includes a third metric based on the segmentation quality of a sclera depicted in the first image.

18. The system of claim 11 , in which the operations further comprise determining a match score by combining, based in part on the quality score, a plurality of match scores, including at least one match score based on the first image.

19. The system of claim 11 , in which the operations further comprise:

determining a match score based on, at least, the first image and data from a reference record that reflects a reference image;

accepting a user based in part on the match score;

comparing the quality score to a previous quality score stored in the reference record; and

updating the reference record with data based on the first image when the quality score is better than the previous quality score.

20. The system of claim 11 , in which the operations further comprise providing feedback based on the quality score to a user.

21. A system, comprising:

a sensor configured to obtain a first image of an eye, wherein the first image includes a view of a portion of a vasculature of the eye external to a corneal limbus boundary of the eye;

a means for determining a plurality of metrics for the first image, wherein the metrics include a first metric for reflecting an extent of one or more connected structures in the first image that represents a morphology of eye vasculature and a second metric for comparing the extent of eye vasculature detected across different color components in the first image by determining a relationship of a variation of a first color difference signal to a variation of a second color difference signal;

a module configured to determine a quality score based on, at least, the plurality of metrics for the first image, wherein the quality score is a prediction of a match score that would be determined based on the first image and a second image, assuming the first image and the second image included a view of the same person's vasculature; and

a module configured to reject or accept the first image based on the quality score.

22. The system of claim 21 , in which determining the first metric comprises:

dilating the vasculature in the first image so that one or more disconnected veins in the vasculature become connected;

thinning the dilated vasculature in the first image; and

determining a portion of the first image that contains the thinned dilated vasculature.

23. The system of claim 21 , wherein determining a relationship of a variation of a first color difference signal to a variation of a second color difference signal, in which determining the second metric comprises:

determining the first color difference signal by subtracting a first color component of the first image from a second color component of the first image;

determining the second color difference signal by subtracting a third color component of the first image from the first color component of the first image; and

calculating a ratio of a standard deviation of the first color difference signal to a standard deviation of the second color difference signal.

24. The system of claim 21 , in which the plurality of metrics includes a third metric based on an amount of glare in the first image.

25. The system of claim 21 , in which the plurality of metrics includes a third metric based on a number and types of occlusions in the first image.

26. The system of claim 21 , in which the plurality of metrics includes a third metric based on a gaze angle of an eye depicted in the first image.

27. The system of claim 21 , in which the plurality of metrics includes a third metric based on the segmentation quality of a sclera depicted in the first image.

28. The system of claim 21 , further comprising:

a module configured to determine a match score by combining, based in part on the quality score, a plurality of match scores, including at least one match score based on the first image.

29. The system of claim 21 , further comprising:

a module configured to determine a match score based on, at least, the first image and data from a reference record that reflects a reference image;

an interface configured to accept a user based in part on the match score;

a module configured to compare the quality score to a previous quality score stored in the reference record; and

a module configured to update the reference record with data based on the first image when the quality score is better than the previous quality score.

30. The system of claim 21 , further comprising:

a user interface configured to provide feedback based on the quality score to a user.