Biometric Authentication System Using Multi-Modal Data Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biometric authentication systems using single types of biometric data face limitations in authentication performance and accuracy, particularly in distinguishing between authorized and unauthorized users, leading to potential false acceptance and rejection rates.
Innovation Solution
A biometric authentication method that determines similarities between multiple types of biometric data (e.g., face and fingerprint) and updates registration data based on these similarities, using a processor to perform authentication and data updates, thereby enhancing recognition accuracy and reducing false acceptance rates by enriching registration data with feature information from multiple biometric sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If single type of biometric data is used for authentication, then device complexity is reduced, but authentication accuracy and reliability deteriorate
Solution Approach 1:
The patent combines multiple types of biometric data (face recognition and fingerprint recognition) into a unified authentication system. The server integrates both biometric modalities to perform joint authentication, where both the face similarity and fingerprint similarity must meet their respective thresholds for successful authentication. This merging approach resolves the contradiction by improving authentication reliability through multi-modal verification while maintaining manageable system complexity through centralized processing architecture.
2Reliability
If multiple types of biometric data are used for authentication, then authentication accuracy is improved, but device complexity increases
Solution Approach 1:
The authentication server is designed with multi-functionality to handle multiple types of biometric data processing. It can simultaneously perform face recognition, fingerprint recognition, and integrated multi-modal authentication. This universal design allows the system to improve authentication accuracy through diverse biometric inputs while avoiding the need for separate dedicated systems for each biometric type, thereby managing overall system complexity.
Solution Approach 2:
The server acts as an intermediary that receives, processes, and integrates multiple biometric data types. It coordinates between different biometric recognition processes, manages the similarity comparisons against registered data, and makes the final authentication decision based on combined results. This intermediary role simplifies the architecture by centralizing complex processing logic in a single coordinating component rather than distributing complexity across multiple independent systems.
3Measurement precision
If registration data is updated frequently with new biometric data, then recognition accuracy is improved, but risk of misregistration increases
Solution Approach 1:
The system implements feedback mechanisms where authentication results and similarity measurements are used to inform subsequent registration updates. When new biometric data is captured, the system compares it against existing registered data, calculates similarity scores, and only updates registration data when authentication succeeds and similarity thresholds are met. This feedback loop ensures that registration data is updated based on verified accurate measurements, improving recognition accuracy while preventing misregistration through conditional update logic.
Solution Approach 2:
The system uses parameter changes in the form of similarity thresholds and reference values to control when registration updates occur. By adjusting these parameters, the system can balance between updating registration data frequently enough to maintain recognition accuracy and being selective enough to avoid misregistration. The dynamic application of threshold parameters allows flexible control over the registration update process based on the quality and reliability of incoming biometric data.
Data Source
AI summary
A biometric authentication method includes determining a first similarity between first biometric data and first registration data; determining a second similarity between second biometric data and second registration data; performing authentication based on the first similarity and the second similarity; and updating the first registration data based on the first biometric data in response to the first similarity being greater than an authentication reference value corresponding to the first similarity and the second similarity being greater than an update reference value corresponding to the second similarity.