Fingerprint Registration Method Reducing False Accept Rate
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fingerprint registration methods require multiple recordings of the same fingerprint, leading to a higher false accept rate and security concerns when multiple fingers are registered, causing inefficiencies and security worries.
Innovation Solution
A fingerprint registration method that receives and processes fingerprint images over multiple inputs, determining feature points to identify matching fingers and adding valid inputs to a candidate collection for template generation, thereby preventing multiple fingers from registering the same fingerprint.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple fingerprint images are recorded to increase registration success rate, then the registration efficiency is improved, but the false accept rate increases and security is compromised
Solution Approach 1:
The patent segments the fingerprint registration process into distinct phases: initial registration phase where multiple images are collected and stored separately, and subsequent verification phase where the judgment module compares new fingerprints against the stored set. This segmentation allows efficient collection of multiple samples while maintaining security through structured verification logic that prevents false accepts.
Solution Approach 2:
The judgment module provides feedback by comparing feature points of newly captured fingerprint images against the stored set of fingerprint images. This feedback mechanism determines whether to accept or reject the registration, enabling the system to maintain high registration success rates while preventing false accepts through intelligent decision-making based on feature point comparison results.
2Ease of operation
If multiple fingers are allowed to register the same fingerprint, then the ease of operation is improved, but the security is compromised
Solution Approach 1:
The system performs preliminary action by storing multiple fingerprint images during the initial registration phase before any verification occurs. This preliminary collection of diverse finger samples enables convenient subsequent operations where users can register with any of their fingers, while the pre-stored images serve as a secure reference set for the judgment module to prevent false accepts.
Solution Approach 2:
The judgment module acts as an intermediary between the fingerprint capture and the registration decision. It mediates by comparing feature points of captured images against the stored set, enabling convenient multi-finger registration while preventing security risks through its filtering and verification function.
3Measurement precision
If feature point comparison is performed for each registration attempt, then the measurement precision is improved, but the processing time increases
Solution Approach 1:
The system applies partial action by performing feature point comparison only on critical characteristics rather than analyzing every detail of the fingerprint image. The judgment module compares key feature points (ridges, valleys, minutiae) to determine matching, achieving sufficient precision for security purposes while reducing processing time by avoiding exhaustive analysis of the entire fingerprint pattern.
Data Source
AI summary
A fingerprint registration method, device and terminal thereof are proposed. The method includes: receiving a registered fingerprint image input at an Nth time by a user and retrieving feature points of the registered fingerprint image input at the Nth time; determining whether the registered fingerprint image input at the Nth time and registered fingerprint images input at N−1 times belong to a same finger based on the feature points of the registered fingerprint image input at the Nth time and feature points of the registered fingerprint images input at the N−1 times previous to the Nth time; and adding the registered fingerprint image input at the Nth time to a candidate collection of registered fingerprint images if the registered fingerprint image input at the Nth time and the registered fingerprint images input at the N−1 times belong to the same finger.


