Curve Segment Matching for Partial Fingerprint Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fingerprint verification systems that rely on minutia processing are ineffective when dealing with partial fingerprint images, as minutia are less prevalent or absent, limiting the use of smaller fingerprint sensors.
Innovation Solution
A system and method for curve segment contour matching that eliminates the need for minutia processing by comparing sets of curve segments from partial fingerprint images, using geometric signatures and curvature angular data to establish correspondence between patterns, allowing for verification without relying on minutia.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If minutia processing is used for fingerprint verification, then verification accuracy is improved, but the system becomes inoperative with partial fingerprint images
Solution Approach 1:
The patent segments the fingerprint image into multiple curve segments instead of relying on discrete minutia points. Each curve segment represents a continuous portion of fingerprint ridges, allowing the system to process partial fingerprint images by matching these segmented curves between reference and test images, thereby maintaining verification accuracy while adapting to reduced image areas.
Solution Approach 2:
The patent inverts the traditional approach by not searching for specific minutia points within partial images, but rather by extracting and matching curve segment features that can be reliably identified even in reduced image areas. This inversion allows the system to work effectively with partial fingerprints by focusing on what can be detected rather than what is traditionally required.
2Area of stationary object
If smaller fingerprint sensors are used, then device size is reduced, but the ability to capture sufficient minutia for verification deteriorates
Solution Approach 1:
The patent divides the fingerprint ridges into multiple curve segments that can be extracted and matched independently. This segmentation allows smaller sensors to capture sufficient verification data by matching multiple curve segments rather than requiring complete minutia sets, thereby maintaining verification reliability while enabling compact sensor design.
Solution Approach 2:
The patent changes the feature extraction parameters from discrete minutia points to continuous curve segment characteristics including curvature, orientation, and geometric signatures. This parameter change allows the system to extract sufficient verification information from smaller sensor areas by utilizing the geometric properties of curve segments rather than relying on the density and distribution of minutia points.
Data Source
AI summary
A system, method, and computer program product for evaluating a conformance of sets of curves using curvature information from curve segments to identify conforming curve segments. These curve segments are evaluated for relative positional variations in a context of a cluster by triangulating corresponding points among the curve segments of the cluster. Little to no deviation between corresponding lengths in the clusters indicate a high probability of match. Confirmation may be done by a more rigorous pattern matcher based upon the conforming curve segments of a cluster.