Systems and methods for privacy-enabled biometric processing
A set of distance measurable encrypted feature vectors can be derived from any biometric data and/or physical or logical user behavioral data, and then using an associated deep neural network (“DNN”) on the output (i.e., biometric feature vector and/or behavioral feature vectors, etc.) an authentication system can determine matches or execute searches on encrypted data. Behavioral or biometric encrypted feature vectors can be stored and/or used in conjunction with respective classifications, or in subsequent comparisons without fear of compromising the original data. In various embodiments, the original behavioral and/or biometric data is discarded responsive to generating the encrypted vectors. In another embodiment, distance measurable or homomorphic encryption enables computations and comparisons on cypher-text without decryption of the encrypted feature vectors. Security of such privacy enabled embeddings can be increased by implementing an assurance factor (e.g., liveness) to establish a submitted credential has not been spoofed or faked.
1. A system for privacy-enabled identification or authentication, the system comprising:
at least one processor operatively connected to a memory;
a classification component executed by the at least one processor, configured to:
generate a first match to an identity label using a first identification or authentication process on at least one encrypted authentication credential generated from a pre-trained neural network as part of an initiating an identification request by an entity seeking to be authenticated or identified;
recover at least one stored encrypted authentication credential stored in memory during prior enrollment, identification, or authentication operations based on using a mapping from the determined identity label from the first match to the identity to the stored encrypted authentication credential, wherein the at least one stored encrypted authentication credential is generated from a pre-trained neural network configured to output the at least one encrypted authentication credential in response to input of unencrypted identification information;
determine a distance on the at least one encrypted authentication credential recovered with the determined identity label and at least one encrypted authentication credential produced during the first identification or authentication process;
return a distance match, responsive to determining the distance between the encrypted authentication credential recovered and the at least one encrypted authentication credential produced meets a threshold for validating a second match to the identity; and
wherein the at least one processor is further configured to generate a liveness score based on a random set of candidate authentication instances.
2. The system of claim 1 , wherein the classification component is further configured to retrieve a group of authentication credentials classified during training based on identification of a group of labels having respective highest values for probability of match.
3. The system of claim 2 , wherein the classification component is further configured to determine an unknown result responsive to the distance determination not meeting the threshold for validating the second match to the identity, or a probability of a match to the first identity not meeting a threshold probability.
4. The system of claim 1 , wherein the at least one processor is configured to build an authentication database including encrypted authentication credentials associated with a respective identification label.
5. The system of claim 1 , wherein the at least one processor is configured to enroll authentication credentials and respective labels for subsequent authentication by a first DNN deep neural network (DNN).
6. The system of claim 5 , further comprising the pre-trained neural network configured to generate the encrypted authentication credentials responsive to unencrypted input of authentication credentials.
7. The system of claim 6 , wherein the at least one processor is configured to associate the encrypted authentication credentials to labels in an authentication database.
8. The system of claim 7 , wherein the at least one processor is configured to validate a plurality of encrypted authentication credentials prior to use in training the first DNN.
9. The system of claim 1 , wherein the at least one processor is configured to reject training instances of authentication credentials that exceed a threshold for validation based on distance evaluation.
10. The system of claim 1 , wherein the at least one processor is further configured to validate contemporaneous input of unencrypted authentication credentials.
11. A computer implemented method for privacy-enabled identification or authentication, the method comprising:
generating, by at least one processor, a first match to an identity label using a first identification or authentication process on at least one encrypted authentication credential generated from a pre-trained neural network as part of an initiating an identification request by an entity seeking to be authenticated or identified;
recovering, by the at least one processor, at least one stored encrypted authentication credential stored in memory during prior enrollment, identification, or authentication operations based on using a mapping from the determined identity label from the first match to the identity to the stored encrypted authentication credential, wherein the at least one stored encrypted authentication credential is generated from a pre-trained neural network configured to output the at least one encrypted authentication credential in response to input of unencrypted identification information;
determining, by the at least one processor, a distance on the at least one encrypted authentication credential recovered with the determined identity label and at least one encrypted authentication credential produced during the first identification or authentication process;
returning, by the at least one processor a distance match, responsive to determining the distance between the encrypted authentication credential recovered and the at least one encrypted authentication credential produced meets a threshold for validating a second match to the identity; and
generating, by the at least one processor, a liveness score based on a random set of candidate authentication instances.
12. The method of claim 11 , wherein the method further comprises retrieving a group of authentication credentials classified during training based on identification of a group of labels having respective highest values for probability of match.
13. The method of claim 12 , wherein the method further comprises determining an unknown result responsive to the distance determination not meeting the threshold for validating the second match to the identity, or a probability of a match to the first identity not meeting a threshold probability.
14. The method of claim 11 , wherein the method further comprises building an authentication database including encrypted authentication credentials associated with a respective identification label.
15. The method of claim 11 , wherein the method further comprises enrolling authentication credentials and respective labels for subsequent authentication by a first deep neural network (DNN).
16. The method of claim 15 , wherein the method further comprises generating the encrypted authentication credentials responsive to unencrypted input of authentication credentials.
17. The method of claim 16 , wherein the method further comprises associating the encrypted authentication credentials to labels in an authentication database.
18. The method of claim 17 , wherein the method further comprises validating a plurality of encrypted authentication credentials prior to use in training the first DNN.
19. The method of claim 11 , wherein the method further comprises rejecting training instances of authentication credentials that exceed a threshold for validation based on distance evaluation.