Systems and methods for privacy-enabled biometric processing

A privacy-enabled biometric system comprising:

• at least one processor operatively connected to a memory;
• an enrollment interface configured to:
• accept unencrypted biometric information;
• generate one or more distance measurable encrypted feature vectors using a first pre-trained neural network from the unencrypted biometric information;
• delete the unencrypted biometric information responsive to generation of the one or more distance measurable encrypted feature vectors by the first pre-trained neural network;
• assign a respective label to link the one or more distance measurable encrypted feature vectors to an identity;
• wherein the first pre-trained neural network is configured to:
• accept unencrypted biometric information from the enrollment interface and an authentication interface;
• encode the unencrypted biometric information into the one or more distance measurable encrypted feature vectors as one way encodings of the unencrypted biometric information;
• and
• the authentication interface configured to:
• process a request to identify or authenticate, the request including respective unencrypted biometric information or a respective one or more distance measurable encrypted feature vectors;
• generate the respective one or more distance measurable encrypted feature vectors using the first pre-trained neural network for the request including the respective unencrypted biometric information;
• and
• determine a match to the identity or an unknown result using the respective one or more distance measurable encrypted feature vectors.

A computer-implemented method for privacy-enabled biometric analysis, the method comprising:

• accepting, by at least one processor, unencrypted biometric information associated with a new entity;
• encoding, by the at least one processor, the unencrypted biometric information into one or more distance measurable encrypted feature vectors that are one way encodings of the unencrypted biometric information using a first pre-trained neural network;
• deleting, by the at least one processor, the unencrypted biometric information responsive to generation of the one or more distance measurable encrypted feature vectors;
• assigning, by the at least one processor, a respective label to link the one or more distance measurable encrypted feature vectors to an identity;
• processing, by the at least one processor, a request to identify or authenticate, the request including respective unencrypted biometric information or a respective one or more distance measurable encrypted feature vectors;
• generating, by the at least one processor, the respective one or more distance measurable encrypted feature vectors using the first pre-trained neural network for the request including the respective unencrypted biometric information;
• and
• determining, by the at least one processor, a match to the identity or an unknown result using the respective one or more distance measurable encrypted feature vectors.