Adaptive Biometric Template Update for Authentication Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in accurately authenticating users due to large intra-class variations in biometric data, such as changes in lighting conditions, facial expressions, and removable accessories, which can lead to errors in recognition and increased vulnerability to impostors.
Innovation Solution
The implementation of user-adaptive biometric authentication systems that adaptively generate and update templates using a threshold-based, gradual learning technique, incorporating incremental-decremental learning to cover a wide range of intra-class variations while maintaining memory efficiency, and utilizing importance scores to optimize enrolled data and manage template storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional biometric authentication systems use fixed templates, then the system is simple to implement, but authentication accuracy deteriorates due to large intra-class variations in biometric data
Solution Approach 1:
The patent implements dynamic template updating where biometric templates are automatically adapted over time based on new input data. The system transitions from static fixed templates to dynamic evolving templates that capture intra-class variations, resolving the contradiction between authentication accuracy and system complexity by making the template structure adaptive rather than fixed
Solution Approach 2:
The system performs self-updating of biometric templates without requiring manual re-enrollment. The automatic template adaptation mechanism allows the system to service itself by continuously learning from new biometric inputs and adjusting templates accordingly, improving accuracy while maintaining operational simplicity
2Measurement precision
If biometric templates are continuously updated to capture intra-class variations, then authentication accuracy improves, but vulnerability to impostors increases
Solution Approach 1:
The patent implements a feedback mechanism where authentication results and similarity scores are used to control template updates. The system only updates templates when confidence thresholds are met, creating a feedback loop that balances accuracy improvement with security maintenance, preventing impostor data from corrupting the template database
Solution Approach 2:
The system dynamically adjusts update parameters such as confidence thresholds and similarity criteria based on authentication outcomes. By changing these parameters adaptively, the system optimizes the balance between capturing legitimate intra-class variations and rejecting impostor attempts, resolving the contradiction between accuracy and security
3Reliability
If multiple biometric templates are stored to cover intra-class variations, then authentication reliability improves, but memory usage increases
Solution Approach 1:
The patent implements an incremental-decremental learning process where outdated or redundant template components are discarded while new informative components are recovered and stored. This selective discarding and recovering mechanism maintains authentication reliability by preserving essential biometric features while removing redundant data, thus optimizing memory usage
Solution Approach 2:
The system extracts only the essential and informative components from biometric data for template storage, rather than storing complete raw templates. By taking out only the critical features that contribute to authentication reliability, the system reduces memory requirements while maintaining performance
4Measurement precision
If biometric authentication requires frequent re-enrollment to maintain accuracy, then template quality improves, but user convenience deteriorates
Solution Approach 1:
The patent implements automatic template adaptation that performs re-enrollment functions without user intervention. The system services itself by continuously updating templates in the background, eliminating the need for users to undergo frequent manual re-enrollment processes, thus maintaining template quality while preserving user convenience
Solution Approach 2:
The system performs preliminary template updates in the background before authentication is needed. By proactively adapting templates during idle periods rather than waiting for authentication failures, the system maintains high template quality without requiring users to experience degradation or perform re-enrollment actions
Data Source
AI summary
Techniques and systems are provided for authenticating a user of a device. For example, input biometric data associated with a person can be obtained. A similarity score for the input biometric data can be determined by comparing the input biometric data to a set of templates that include reference biometric data associated with the user. The similarity score can be compared to an authentication threshold. The person is authenticated as the user when the similarity score is greater than the authentication threshold. The similarity score can also be compared to a learning threshold that is greater than the authentication threshold. A new template including features of the input biometric data is saved for the user when the similarity score is less than the learning threshold and greater than the authentication threshold.