Systems and methods for privacy-enabled biometric processing
In one embodiment, a set of feature vectors can be derived from any biometric data, and then using a deep neural network (“DNN”) on those one-way homomorphic encryptions (i.e., each biometrics' feature vector) can determine matches or execute searches on encrypted data. Each biometrics' feature vector can then be stored and/or used in conjunction with respective classifications, for use in subsequent comparisons without fear of compromising the original biometric data. In various embodiments, the original biometric data is discarded responsive to generating the encrypted values. In another embodiment, the homomorphic encryption enables computations and comparisons on cypher text without decryption. This improves security over conventional approaches. Searching biometrics in the clear on any system, represents a significant security vulnerability. In various examples described herein, only the one-way encrypted biometric data is available on a given device. Various embodiments restrict execution to occur on encrypted biometrics for any matching or searching.
1. A privacy-enabled biometric system comprising:
at least one processor operatively connected to a memory;
a classification component executed by the at least one processor, comprising a classification network having a deep neural network configured to:
classify Euclidean measurable encrypted feature vectors and label inputs for identification during training, and
accept as an input, to the deep neural network, encrypted feature vectors that are Euclidean measurable and are produced as a one way encoding of unencrypted biometric information by a first pre-trained neural network, and return an unknown result or a matching label as output; and
an enrollment interface configured to:
accept unencrypted biometric information;
provide Euclidean measurable encrypted feature vectors generated from the first pre-trained neural network based on the unencrypted biometric information;
delete the unencrypted biometric information responsive to generation of the Euclidean measurable encrypted feature vectors by the first pre-trained neural network;
provide a respective label to the classification component for training; and
trigger the classification component to train the deep neural network on the Euclidean measurable encrypted feature vectors to link the label to respective encrypted feature vectors;
wherein the first pre-trained neural network is configured to:
accept unencrypted biometric information; and
encode the unencrypted biometric information into Euclidean measurable encrypted feature vectors as one way encodings of the unencrypted biometric information.
2. The system of claim 1 , wherein the enrollment interface is accessible via uri, and is configured to accept unencrypted biometric information and personally identifiable information (“PII”).
3. The system of claim 2 , wherein the enrollment interface is configured to link the PII to a one way homomorphic encryption of an unencrypted biometric input.
4. The system of claim 1 , wherein the classification component is further configured to return a set of probabilities for matching an input Euclidean measurable encrypted feature vector.
5. The system of claim 1 , wherein the classification component is further configured to predict an outcome based on a trained model, a set of inputs for the prediction and a result of a match or unknown.
6. The system of claim 1 , further comprising an interface configured to accept a biometric input and return an indication of known or unknown to a requesting entity.
7. The system of claim 6 , wherein the requesting entity includes any one or more of: an application, a mobile application, a local process, a remote process, a method, and a business object.
8. The system of claim 1 , wherein the classification component further comprises multiple classification networks for processing different types of biometric information.
9. The system of claim 1 , wherein the classification component is further configured to match an identity of a person responsive to at least two probable biometric indicators, wherein the at least two probable biometric indicators are evaluated as part of a voting algorithm.
10. The system of claim 1 , wherein the system is further configured to enroll an entity for biometric authentication; and the classification component is further configured to accept input of Euclidean measurable encrypted feature vectors for person identification during prediction.
11. A computer implemented method for privacy-enabled biometric analysis, the method comprising:
instantiating, by at least one processor, a classification component comprising a deep neural network configured to:
classify encrypted feature vectors that are Euclidean measurable and label inputs for identification during training, and
accept as an input encrypted feature vectors that are Euclidean measurable and are produced as a one way encoding of unencrypted biometric information by a first pre-trained neural network, and return an unknown result or a label as output during prediction:
instantiating, by the at least one processor, an enrollment interface;
accepting, by the enrollment interface, unencrypted biometric information associated with a new entity;
encoding, by the first pre-trained neural network, the unencrypted biometric information into Euclidean measurable encrypted feature vectors that are one way encodings of the unencrypted biometric information;
deleting, by the at least one processor, the unencrypted biometric information responsive to generation of the Euclidean measurable encrypted feature vectors by the first pre-trained neural network;
triggering, by the at least one processor, the classification component to train the deep neural network on the Euclidean measurable encrypted feature vectors derived from the biometric information to link a respective label to respective encrypted feature vectors; and
returning, by the classification component, an unknown result or a matching label for subsequent identification.
12. The method of claim 11 , wherein instantiating the enrollment interface includes hosting a portal accessible via uri, and the method includes accepting biometric information and personally identifiable information (“PII”) through the portal.
13. The method of claim 11 , wherein the method further comprises linking personally identifiable information (“PII”) to a one way homomorphic encryption of an unencrypted biometric input.
14. The method of claim 11 , wherein the method further comprises:
enrolling individuals for biometric authentication; and
mapping labels and respective feature vectors for person identification, responsive to input of encrypted Euclidean measurable feature vectors and a label for the individual.
15. The method of claim 11 , wherein the method further comprises returning a set of probabilities for matching a set of existing labels.
16. The method of claim 15 , wherein the method further comprises predicting an outcome based on a trained model, a set of inputs for the prediction and a result of a class or unknown.
17. The method of claim 11 , further comprising accepting via an authentication interface a biometric input and returning an indication of known or unknown to a requesting entity.
18. The method of claim 17 , wherein the requesting entity includes any one or more of: an application, a mobile application, a local process, a remote process, a method, and a business object.
19. The method of claim 11 , wherein the method further comprises processing different types of biometric information using multiple classification networks.
20. The method of claim 11 , wherein the method further comprises generating an identity of a person responsive to at least two probable biometric indicators, wherein the at least two probable biometric indicators are evaluated as part of a voting algorithm.