Biometric Code Generation via Feature Normalization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional encryption systems face security risks due to the need to store private keys, and biometric security systems are vulnerable to unauthorized access of user profiles, which can compromise data integrity and user privacy.

Innovation Solution

A method is developed to generate a unique code from biometric samples by analyzing feature values, normalizing them, calculating mean and variance, setting quantization levels, de-correlating the values, and combining them to create a subset of bits that can be used as an encryption key, eliminating the need for storing biometric templates and enhancing security by associating the key with the biometric sample.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If private keys are stored in conventional encryption systems, then key retrieval and exchange during communications is enabled, but security is compromised due to potential unauthorized access to stored keys

Engineering Contradiction:

Improvekey retrievalVSAvoidsecurity

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The invention extracts only the necessary feature values from biometric samples and uses these to generate encryption keys dynamically, eliminating the need to store complete biometric templates or private keys. The system takes out only the essential information needed for key generation while discarding sensitive data that would require secure storage.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system enables self-service by generating encryption keys directly from biometric samples without requiring external key management infrastructure. The biometric sample itself serves as the source for key generation, eliminating the need for separate key storage and retrieval mechanisms.

Inventive Principle:

Principle #25Self-service

2Reliability

If biometric templates are stored in a database, then user verification is enabled, but unauthorized access to the database can compromise data integrity and user privacy

Engineering Contradiction:

Improveuser verificationVSAvoidunauthorized access risk

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system extracts only the necessary feature values from biometric samples and uses these to generate encryption keys dynamically, eliminating the need to store complete biometric templates or private keys. The system takes out only the essential information needed for key generation while discarding sensitive data that would require secure storage.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The invention introduces feature values and encryption algorithms as intermediaries between the biometric sample and the verification process. Instead of storing and comparing raw biometric templates, the system uses derived feature values to generate keys that serve as the actual verification mechanism, protecting the original biometric data.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If complete biometric templates are stored for security verification, then accurate user authentication is achieved, but user reluctance increases due to privacy concerns about storing personal data

Engineering Contradiction:

Improveauthentication accuracyVSAvoiduser acceptance

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The invention changes the parameter of data storage from complete biometric templates to derived feature values and encryption keys. This parameter change maintains authentication functionality while significantly reducing the privacy concerns associated with storing sensitive personal biometric data.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS8165289B2Method and apparatus for the generation of code from pattern features

Publication Date: 2012.04.24 UNIVERSITY OF KENT

AI summary

A method is provided for deriving a single code from a biometric sample in a way which enables different samples of a user to provide the same code whilst also distinguishing between samples of different users. Different features are analysed to obtain mean and variance values, and these are used to control how the different feature values are interpreted. In addition, features are combined and a sub-set of bits of the combination is used as the code. This enables bits which are common to all user samples to be dropped as well as bits which may differ between different samples of the same user.