Biometric Template Matching on Smart Cards
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric template matching in resource-constrained environments, such as smart cards, faces challenges due to the need for complex alignment and comparison of minutiae features, which requires significant memory and computational resources, and often involves trigonometric functions not available in standard smart cards.
Innovation Solution
A biometric template matching method that restricts the allowed rotation between two templates, comparing orientation data and determining displacement vectors to assess matching, thereby avoiding the need for rotation and trigonometric calculations, and using a coarser matching process within the smart card and a finer matching process in a connected device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex alignment and comparison of minutiae features is performed in smart cards, then biometric template matching accuracy is improved, but memory and computational resources are insufficient
Solution Approach 1:
The patent divides the biometric authentication system into two segments: the smart card handles initial template storage and basic matching operations, while a connected device (PC, smartphone) performs complex alignment and minutiae comparison. This segmentation allows each component to operate within its resource constraints while achieving high matching accuracy through the connected device's computational power.
Solution Approach 2:
The patent introduces a connected device as an intermediary between the smart card and the biometric matching process. The smart card communicates biometric templates to the connected device, which then performs the computationally intensive alignment and comparison operations, acting as a mediator that bridges the resource constraints of the smart card with the accuracy requirements of biometric matching.
2Measurement precision
If rotation and translation alignment of templates is performed, then matching accuracy is improved, but computational burden and memory requirements increase
Solution Approach 1:
The patent performs preliminary actions by storing normalized biometric templates in the smart card during the enrollment phase. The templates are pre-processed to establish reference orientations, so that during authentication, only minimal alignment adjustments are needed rather than performing full rotation and translation calculations from scratch.
Solution Approach 2:
The patent changes the parameter representation by using discrete orientation angles (0°, 45°, 90°, 135°) instead of continuous rotation parameters. This discretization reduces the computational complexity of alignment operations, allowing the system to achieve sufficient matching accuracy with fewer computational resources by comparing templates at these predefined angular intervals.
3Measurement precision
If trigonometric functions are used for alignment calculations, then matching precision is improved, but smart card compatibility is reduced
Solution Approach 1:
The patent uses simplified alignment methods that do not require trigonometric functions, accepting slightly lower precision in exchange for broad smart card compatibility. The system uses basic geometric comparisons and discrete angle matching that can be implemented on any standard smart card processor, sacrificing some alignment precision to ensure universal compatibility across different smart card implementations.
4Speed
If central database of fingerprints is maintained, then identification speed is improved, but security against fraud is reduced
Solution Approach 1:
The patent extracts the biometric template from the central database and stores it in a portable smart card. This extraction removes the vulnerability of centralized storage while maintaining fast identification performance, as the template is readily available in the smart card for immediate comparison during authentication without requiring database queries.
Solution Approach 2:
The patent creates a copy of the biometric template in the portable smart card that can be used for authentication without exposing the original template in the central database. This copying approach enables fast local matching while maintaining security, as the central database remains protected and only the copied template in the smart card is used for authentication operations.
Data Source
AI summary
A biometrics template matching method includes the steps of: providing a reference biometric template and a candidate biometric template, each including position data and orientation data of a respective plurality of minutiae; comparing the orientation data of each minutia from the candidate template with the orientation data of each minutia from the reference template; when the orientation data of a selected pair differ by no more than a first threshold, determining a displacement vector representative of the difference in position data of the selected pair of minutiae; determining the maximum number of displacement vectors that differ from each other by less than a second threshold; if the maximum number of displacement vectors is less than a third threshold, returning a mismatch, otherwise returning a match.