Biometric template security and key generation

1. A computer-implemented method comprising:

receiving one or more images;

identifying a plurality of interest points based on the received images;

associating one or more real descriptors with each interest point, wherein each real descriptor describes one or more localities surrounding the corresponding interest point;

scrambling one or more of the real descriptors using a device-specific signature;

generating a plurality of obfuscating data points based on the interest points;

generating and encoding a key using a subset of at least one of the obfuscating data points and the interest points;

creating an obfuscated template based on the interest points, the obfuscating data points, and the device-specific signature; and

storing the obfuscated template.

2. The method of claim 1 , wherein the obfuscating data points are generated such that a spatial distribution of the interest points and a spatial distribution of the obfuscating data points are substantially similar.

3. The method of claim 1 , further comprising discarding a record of which points in the obfuscated template are the interest points.

4. The method of claim 1 , wherein one or more of the obfuscating points are tagged by one or more of the interest points, and generating and encoding the key comprises using a subset of at least one of the tagged obfuscating data points and the interest points.

5. The method of claim 4 , wherein one or more of the interest points are tagged by one or more of the interest points, and generating and encoding the key comprises using a subset of at least one of the tagged obfuscating data points and the tagged interest points.

6. The method of claim 5 , wherein a number of the tagged obfuscating points is greater than a number of the tagged interest points.

7. The method of claim 1 , wherein each point in the subset is determined based on a different one of the interest points.

8. The method of claim 1 , wherein the images comprise biometric imagery.

9. The method of claim 8 , wherein the images comprise images of a region of an eye, each eye region image comprising a view of a vasculature of the respective eye region, and wherein the interest points comprise vascular interest points.

10. The method of claim 1 , further comprising associating one or more synthesized descriptors with each obfuscating data point, wherein each synthesized descriptor comprises a statistical similarity to the real descriptors.

11. The method of claim 10 , further comprising:

receiving one or more second images;

identifying a second plurality of interest points based on the received second images;

creating a verification template based on the second plurality of interest points;

comparing the verification template with the obfuscated template to identify a plurality of matching interest points; and

authenticating a user based on the matching interest points.

12. The method of claim 11 , wherein the comparing comprises identifying the matching interest points based on one or more of the real and synthesized descriptors.

13. The method of claim 11 , further comprising reducing a dimensionality of the real descriptors and the synthesized descriptors.

14. The method of claim 11 , wherein the comparing comprises identifying the matching interest points based on one or more of the reduced dimensionality descriptors.

15. The method of claim 11 , wherein the scrambling comprises isometrically scrambling one or more of the real descriptors and one or more of the synthesized descriptors using the device-specific signature.

16. The method of claim 15 , wherein the comparing comprises identifying the matching interest points based on one or more of the scrambled descriptors.

17. The method of claim 11 , further comprising decoding a key based on at least a subset of the matching interest points.

18. The method of claim 1 , wherein each interest point is located in a respective tile of a plurality of neighboring tiles.

19. A system comprising:

one or more computers programmed to perform operations comprising:

receiving one or more images;

identifying a plurality of interest points based on the received images;

associating one or more real descriptors with each interest point, wherein each real descriptor describes one or more localities surrounding the corresponding interest point;

scrambling the one or more of the real descriptors using a device-specific signature;

generating a plurality of obfuscating data points based on the interest points;

generating and encoding a key using a subset of at least one of the obfuscating data points and the interest points;

creating an obfuscated template based on the interest points, the obfuscating data points, and a device-specific signature; and

storing the obfuscated template.

20. The system of claim 19 , wherein the obfuscating data points are generated such that a spatial distribution of the interest points and a spatial distribution of the obfuscating data points are substantially similar.

21. The system of claim 19 , wherein the operations further comprise discarding a record of which points in the obfuscated template are the interest points.

22. The system of claim 19 , wherein one or more of the obfuscating points are tagged by one or more of the interest points, and generating and encoding the key comprises using a subset of at least one of the tagged obfuscating data points and the interest points.

23. The system of claim 22 , wherein one or more of the interest points are tagged by one or more of the interest points, and generating and encoding the key comprises using a subset of at least one of the tagged obfuscating data points and the tagged interest points.

24. The system of claim 23 , wherein a number of the tagged obfuscating points is greater than a number of the tagged interest points.

25. The system of claim 19 , wherein each point in the subset is determined based on a different one of the interest points.

26. The system of claim 19 , wherein the images comprise biometric imagery.

27. The system of claim 26 , wherein the images comprise images of a region of an eye, each eye region image comprising a view of a vasculature of the respective eye region, and wherein the interest points comprise vascular interest points.

28. The system of claim 19 , wherein the operations further comprise associating one or more synthesized descriptors with each obfuscating data point, wherein each synthesized descriptor comprises a statistical similarity to the real descriptors.

29. The system of claim 28 , wherein the operations further comprise:

receiving one or more second images;

identifying a second plurality of interest points based on the received second images;

creating a verification template based on the second plurality of interest points;

comparing the verification template with the obfuscated template to identify a plurality of matching interest points; and

authenticating a user based on the matching interest points.

30. The system of claim 29 , wherein the comparing comprises identifying the matching interest points based on one or more of the real and synthesized descriptors.

31. The system of claim 29 , wherein the operations further comprise reducing a dimensionality of the real descriptors and the synthesized descriptors.

32. The system of claim 29 , wherein the comparing comprises identifying the matching interest points based on one or more of the reduced dimensionality descriptors.

33. The system of claim 29 , wherein the scrambling comprises isometrically scrambling one or more the real descriptors and one or more of the synthesized descriptors using the device-specific signature.

34. The system of claim 33 , wherein the comparing comprises identifying the matching interest points based on one or more of the scrambled descriptors.

35. The system of claim 29 , wherein the operations further comprise decoding a key based on at least a subset of the matching interest points.

36. The system of claim 19 , wherein each interest point is located in a respective tile of a plurality of neighboring tiles.