Multi-view Fingerprint Matching via Composite Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric recognition systems face challenges in reliably recognizing fingerprints from partial views due to reduced discriminative information and inaccuracies in aligning multiple enrollment views, leading to false matches or non-matches.
Innovation Solution
A method and device for biometric matching that acquire a verification template, compare it to individual enrollment views, calculate a composite match score based on individual and inter-view match statistics, and indicate a match if the score satisfies a threshold, improving recognition reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple enrollment views are stitched together into a single larger view, then the enrollment template provides more comprehensive fingerprint information, but alignment inaccuracies produce distortions at boundaries that detrimentally impact match performance
Solution Approach 1:
The patent divides the enrollment template into multiple separate enrollment views instead of stitching them into a single view. Each enrollment view is kept as an independent segment that can be individually compared with the verification view, avoiding the alignment distortion problems that arise from stitching multiple views together.
Solution Approach 2:
The patent introduces geometric relationships as an intermediary constraint between enrollment views and verification views. These geometric relationships constrain the alignment transformations, ensuring that comparisons between views are performed with proper spatial constraints that account for the sensor's field of view and the finger's geometry.
2Device complexity
If individual enrollment views are compared separately with simple score combination, then the matching process is computationally simpler, but false matches occur when imposters match strongly with only a single view
Solution Approach 1:
The patent implements a feedback mechanism where geometric relationships between enrollment views provide constraints that inform the matching process. The geometric constraints feedback to the score combination process, allowing the system to weight and combine individual view scores in a way that accounts for the spatial relationships between views, thereby reducing false matches.
Solution Approach 2:
The patent changes the parameters used in score combination from simple additive or average scoring to a more sophisticated scoring function that incorporates geometric relationship parameters. The match score is computed as a function of individual view match scores and inter-view match scores, with weights and thresholds adjusted based on geometric constraints.
3Volume of moving object
If the sensor size is reduced for portable applications, then the device becomes more portable, but the sensor can only capture partial views of the fingerprint
Solution Approach 1:
The patent transitions from a single 2D fingerprint view to multiple 2D views that collectively represent a more complete 3D understanding of the fingerprint surface. By capturing multiple partial views from different positions or orientations and comparing them against multiple enrollment views, the system compensates for the limited field of view of small sensors.
Solution Approach 2:
The patent segments both the verification process and enrollment template into multiple view comparisons. Instead of requiring a single complete fingerprint view, the system divides the matching process into multiple independent view comparisons, each using a portion of the fingerprint, and combines the results to achieve reliable authentication.
Data Source
AI summary
A method and a device are provided for performing a recognition process. The recognition process compares an individual fingerprint view to a fingerprint enrollment template in order to determine whether a match has been found. The determination of a match is based on individual match statistics collected between the individual fingerprint view and each view of the fingerprint enrollment template. Additionally, inter-view match statistics between each view of the fingerprint enrollment template may also be determined. The inter-view match statistics can be analyzed along with the individual match statistics to further inform the determination of a match between the individual fingerprint view and the fingerprint enrollment template.