US 11,775,631 B1Granted

System and method for perfecting and accelerating biometric identification via evolutionary biometrics via continual registration

AssigneeKnwn Technologies, Inc.

InventorRichard Lawrence Kane

Life of this patent

Estimating the term…

Loading patent…

Abstract

A system and method for automatically updating biometric data for a user by preferably retaining an ongoing digital collection of images/biometric captures/IOT (“Internet of Things”) captures taken over the life of the user. Changes in appearance of the user due to age, health, facial hair, hair color, hair length and many other evolutionary changes are automatically captured and contribute to the learning set. Accordingly, the system itself gets smarter and faster with every subsequent authentication. In one non-limiting embodiment, a target of the most recent 500 captures are retained, and tuning may suggest a greater number of captures should or could be used. Preferably, with each new capture the oldest capture in the ongoing digital collection is automatically deleted by the system.

Claims12

Dependent claims

A software-based method for evolving a biometric signature for an individual with each subsequent validation for use in identifying or authenticating the individual for a transaction or other activity in order to decrease the processing time required for identifying or authenticating the individual and to improve the accuracy of the identification or authentication, comprising the steps of:

(a) continually and passively receiving or capturing current digital data from an electronic device possessed by an individual by an electronic system prior to authenticating the individual for a transaction, wherein the current digital data includes Internet of Things data;
wherein the Internet of Things data is passively obtained from electronic devices around or in range of the individual without regard to whether the electronic devices are registered or not with the electronic system;
(b) adding the current digital data to a stack of previously stored digital data for the individual;
(b1) creating an updated stack of digital data using the current digital data and the stack of previously stored digital data;
(b2) evolving a biometric signature for the individual using the updated stack of digital data;
(c) storing the updated stack to one or more electronic storage devices;
(d) where a number of digital data captures for the individual previously received or captured for the stack has reached a previously determined or programmed maximum number of digital data captures, electronically deleting an oldest stored digital data capture from the stack so that the updated stack does not exceed the maximum number of digital data captures with the addition of the current digital data in step (b);
(e) causing a neural network component of the electronic system to relearn and evolve the biometric signature for the individual each time the stack of digital data for the individual is updated;
and
(f) anticipating behavioral patterns for the individual based on the updated stack of digital data.

7 more elements

(a) continually and passively receiving or capturing current digital data from an electronic device possessed by an individual by an electronic system prior to authenticating the individual for a transaction, wherein the current digital data includes Internet of Things data;
wherein the Internet of Things data is passively obtained from electronic devices around or in range of the individual without regard to whether the electronic devices are registered or not with the electronic system;
(b) adding the current digital data to a stack of previously stored digital data for the individual;
(c) creating an updated stack of digital data using the current digital data and the stack of previously stored digital data;
(d) evolving a biometric signature for the individual using the updated stack of digital data;
(e) storing the updated stack to one or more electronic storage devices;
(f) where a number of digital data captures for the individual previously received or captured for the stack has reached a previously determined or programmed maximum number of digital data captures, electronically deleting an oldest stored digital data capture from the stack so that the updated stack does not exceed the maximum number of digital data captures with the addition of the current digital data;
(g) causing a neural network component of the electronic system to relearn and evolve the biometric signature for the individual each time the stack of digital data for the individual is updated;
(h) anticipating behavioral patterns for the individual based on the updated stack of digital data;
and
(i) prefetching the previously stored digital data for the individual using the Internet of Things data.

8 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/590,285Filed 1 Oct 2019 · us provisional application
Priority: 2 Oct 2018
Granted: 3 Oct 2023
Term ends: Not estimatedThe source records don’t support an estimate.

Classification8 codes

G06F 21/45
Authentication, i.e. establishing the identity or authorisation of security principals: structures or tools for the administration of authentication
G06N 3/04
Neural networks: architecture, e.g. interconnection topology
G06N 3/096
Learning methods: transfer learning
G06N 20/00
Machine learning