Biometric Security Token Verification via FEHash Templates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional security systems relying on complex security tokens are difficult for users to remember, making them predictable, and biometric encryption methods pose risks due to unchangeable biometric information, lacking the ability to change or cancel encryption keys without storing personal data.
Innovation Solution
A system and method that verifies users through a security token combined with biometric information processing, using Full Entropy Hash (FEHash) to generate and compare cryptographic templates without storing biometric data, allowing for key changes and cancellations by applying Random Fourier Feature Mapping and Linear Support Vector Machine techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If complex security tokens are used to improve security strength, then reliability is improved, but ease of operation deteriorates because users cannot remember them
Solution Approach 1:
The patent introduces biometric information as an intermediary that bridges the gap between security requirements and user convenience. Instead of directly using complex tokens that users must remember, the system uses biometric data (fingerprint, face, iris) as a mediator to automatically generate and verify security tokens, eliminating the need for users to memorize complex sequences while maintaining strong security through cryptographic hashing functions.
2Reliability
If biometric information is stored to enable encryption, then reliability is improved, but harmful factors increase due to exposure risks and inability to change encryption keys
Solution Approach 1:
The patent extracts only the essential cryptographic elements (hash values and comparison results) from the biometric verification process, while deliberately excluding the storage of actual biometric data. The system extracts feature information from biometric inputs, processes it through cryptographic hashing, and stores only the resulting hash values and decision hyperplane parameters, completely removing the harmful element of storing sensitive biometric information while preserving the useful encryption and verification capabilities.
Solution Approach 2:
The patent transforms the biometric verification system by changing the parameter being stored from raw biometric data to cryptographic hash values. This parameter transformation allows the system to maintain reliability for encryption and verification purposes while eliminating the security risks associated with storing actual biometric information, as hash values are one-way transformations that cannot be reversed to obtain original biometric data.
3Reliability
If biometric information is stored to enable verification, then reliability is improved, but loss of information increases when biometric data must be protected or deleted
Solution Approach 1:
The patent extracts and stores only the verification results (hash comparison outcomes) and cryptographic parameters needed for future verification, while excluding the storage of actual biometric information. This extraction approach maintains verification accuracy by preserving the essential cryptographic elements (hash values and decision hyperplane parameters) while eliminating the need to protect and manage sensitive biometric data, thereby reducing information protection overhead.
Data Source
AI summary
A user verifying method comprises: receiving an enrollment request signal including biometric information for enrollment from a user terminal; extracting feature information from the biometric information; embedding the feature information extracted from the biometric information; generating a projection parameter and a decision hyperplane parameter based on the embedding result; storing the projection parameter and the decision hyperplane parameter; generating a first template by applying a cryptographic hashing function to a pre-generated first key; receiving a verification request signal including biometric information from the user terminal; extracting feature information from the biometric information; embedding the feature information extracted from the biometric information; generating a second key corresponding to the first key based on the embedding result; generating a second template by applying the cryptographic hashing function to the second key; comparing the first template with the second template; and completing user verification based on the comparison result.