Fingerprint Matching Integrity Check Using Redundant Score Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems are vulnerable to attacks that manipulate the matching score between candidate and reference fingerprints, leading to unauthorized access.
Innovation Solution
A method involving Delaunay triangulation of minutiae to generate local features, computing similarity and geometric coherence scores, and performing redundant matching score calculations to detect tampering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single matching score calculation is performed, then the processing time and computational resources are reduced, but the system becomes vulnerable to tampering attacks that manipulate the matching score
Solution Approach 1:
The system performs preliminary calculations of geometric coherence scores and stores them during the first matching process. These pre-computed values are then used in the second matching process to verify consistency, avoiding the need to recompute all geometric relationships from scratch and thus reducing the time penalty of redundant verification.
Solution Approach 2:
The system uses the first matching score and geometric coherence scores as feedback to verify the integrity of the second matching score. By comparing geometric coherence scores between the two matching processes, the system can detect tampering attempts and reject fraudulent authentication, thereby improving security without requiring complete reprocessing.
2Reliability
If redundant matching score calculations are performed to detect tampering, then the security against manipulation attacks is improved, but the computational complexity and processing time increase
Solution Approach 1:
The system segments the fingerprint matching process into two distinct phases: a first matching process that computes both similarity scores and geometric coherence scores, and a second matching process that verifies consistency using the pre-computed geometric coherence scores. This segmentation allows the system to perform redundant verification only for the critical matching score calculation, rather than duplicating the entire complex matching algorithm, thus reducing overall computational complexity.
Solution Approach 2:
Instead of performing complete redundant matching calculations, the system performs a partial verification action by only recomputing and comparing the matching scores while utilizing the pre-computed geometric coherence scores from the first process. This partial action provides sufficient security verification without the full computational burden of complete redundancy.
3Reliability
If geometric coherence scores are computed and stored during the first matching process, then the ability to detect tampering is improved, but the memory requirements and processing overhead increase
Solution Approach 1:
The system extracts only the essential geometric coherence scores from the first matching process and stores them for verification in the second process. By taking out only the critical verification data rather than storing all intermediate calculation results, the system minimizes memory requirements while maintaining the ability to detect tampering attempts.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
The present invention relates to a method for determining a match between a candidate fingerprint and a reference fingerprint characterized by minutiae local features comprising : - extracting (S1) several minutiae from the candidate fingerprint, - computing (S2) from said extracted minutiae a plurality of minutiae local features of the candidate fingerprint, - for at least one local feature pair, called current local feature pair, among a plurality of local feature pairs comprising each a candidate fingerprint local feature and a reference fingerprint local feature: - computing (S3) a first similarity score evaluating the similarity of the candidate fingerprint local feature and the reference fingerprint local feature of the current local feature pair, and storing (S4) said computed first similarity score, - determining (S5) a match between the candidate fingerprint local feature and the reference fingerprint local feature of the current local feature pair based on said computed first similarity score, - computing (S6) a first global matching score between the candidate fingerprint and the reference fingerprint based on the first similarity scores of said matching local feature pairs; - for each matching local feature pair, computing (S7) a second similarity score evaluating the similarity of the candidate fingerprint local feature and the reference fingerprint local feature of the matching local feature pair, - computing (S8) a second global matching score between the candidate fingerprint and the reference fingerprint based on said computed second similarity scores, - determining a match (S9) between the candidate fingerprint and the reference fingerprint comprising: comparing the first and second global matching scores and, comparing the first matching score to a matching threshold.