Systems and methods to verify identity of an authenticated user using a digital health passport
Inventors
SLY, Axel • SHARMA, Srivatsa Akshay • REDINGER, Brett Robert • REICH, Devin Daniel • Trooskens, Geert • LOOTUS, Meelis • Lee, Young Jin • ARREDONDO, Ricardo Lopez • KAUTZ, IV, Frederick Franklin • BHAT, Satish Srinivasan • KIRK, Scott Michael • DE BROUWER, Walter Adolf • THAKORE, Kartik
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11177960-B2
Publication Date
2021-11-16
Expiration Date
Abstract
The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.
Core Innovation
The invention relates to a digital health passport for return-to-work authorization without relying on a central identity/health database, using authentication of an authenticated user to generate a scannable code for a validator to scan. The method securely sends, to a registration server, a unique identifier for the authenticated user, a cryptographic signature generated based at least in part upon a biometric identifier for the authenticated user, and a timestamp.
The registration server provides a success nonce usable by the registration server to access a user photo identifier and access a hash of at least the signature. The method recreates the hash of the at least the signature stored by the registration server and generates the scannable code by encrypting the success nonce and the recreated hash, and the encrypted contents are produced into the scannable code and displayed for a validator to scan for verifying that a return-to-work request is from the authenticated user.
The biometric identifier of the authenticated user is generated by feeding a plurality of non-deterministic biometric inputs, including a plurality of face images and a plurality of voice samples of a user, to a trained machine learning model producing a plurality of feature vectors. The feature vectors are projected onto a surface of a unit hyper-sphere and a characteristic identity vector representing the user is computed based on the feature vectors as projected. Authentication and verification are tied to the cryptographic signature, the recreated hash, and the nonce code carried by the scannable code.
Claims Coverage
The patent includes three independent claims: a method of generating a scannable code for an authenticated user, a registration-server method of verifying a return-to-work request based on a scanned scannable code, and a non-transitory computer-readable storage medium implementing the code-generation method. Across these independent claims, the main inventive features combine biometric identifier generation from non-deterministic face and voice inputs using a trained machine learning model with feature-vector projection onto a unit hyper-sphere, and cryptographic protection of a one-time success nonce and a recreated hash within a scannable code for validator scanning and registration-server verification.
Scannable code generation using encrypted success nonce and recreated hash
securely sending, to a registration server, a unique identifier for the authenticated user, a cryptographic signature generated based at least in part upon a biometric identifier for the authenticated user, and a timestamp; receiving, from the registration server, a success nonce usable by the registration server to access a user photo identifier and access a hash of at least the signature; recreating the hash of the at least the signature stored by the registration server; generating the scannable code including encrypting the success nonce and the recreated hash and producing the scannable code from results of the encrypting; and displaying the scannable code for a validator to scan for verifying that a return-to-work request is from the authenticated user.
Registration-server verification of return-to-work request via scanned nonce code and recreated hash
securely receiving, from a validator that has scanned a scannable code that encodes a nonce code, previously generated by the registration server, and a recreated hash, previously associated by the registration server with the nonce; extracting the nonce code and the recreated hash from the scannable code; using the nonce code to access a hash of record and photo ID of the authenticated user; validating the recreated hash against a hash of record; marking the nonce code as used; and sending the validator, a photo of the authenticated user and a success message to return-to-work.
Biometric identifier generation using unit hyper-sphere projected feature vectors
wherein the biometric identifier of the authenticated user was generated by feeding a plurality of non-deterministic biometric inputs, including a plurality of face images and a plurality of voice samples of a user, to a trained machine learning model producing a plurality of feature vectors, projecting the plurality of feature vectors onto a surface of a unit hyper-sphere and computing a characteristic identity vector representing the user based on the feature vectors as projected.
Overall claim coverage centers on generating a scannable code for an authenticated user by encrypting a success nonce together with a recreated hash, displaying it for validator scanning, verifying return-to-work requests at the registration server by extracting and validating nonce code and recreated hash, and generating the biometric identifier from non-deterministic face images and voice samples using a trained machine learning model with feature-vector projection onto a unit hyper-sphere to compute a characteristic identity vector.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?