Fingerprint Matching Method Using True Feature Point Filtering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fingerprint matching methods are significantly affected by pseudo feature points, leading to a reduced success rate, especially in low-quality images or those with weak preprocessing.

Innovation Solution

A fingerprint matching method that extracts feature points, performs an initial matching to identify true feature points by calculating direction differences, removes falsely matched pairs based on distance and frequency criteria, and conducts a second matching using the remaining true feature point pairs, thereby improving the accuracy and success rate.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing fingerprint matching methods use information from multiple surrounding feature points for matching, then the matching process can be performed, but the success rate is significantly reduced due to pseudo feature points

Engineering Contradiction:
Improvesuccess rate of fingerprint matchingVSAvoidimpact of pseudo feature points
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent performs preliminary identification of true feature points before conducting the final matching process. By calculating direction differences between candidate feature points and their surrounding points, and verifying geometric relationships in advance, the system filters out pseudo feature points before they can negatively impact the matching success rate

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements a feedback mechanism where the matching process uses direction difference calculations and geometric relationship verification to continuously evaluate and refine feature point identification. The system adjusts its selection of feature points based on whether they satisfy predetermined conditions regarding direction consistency and geometric relationships with surrounding points

Inventive Principle:
Principle #23Feedback

2Ease of manufacture

If the fingerprint image quality is low or preprocessing is weak, then image acquisition is easier, but the matching accuracy is significantly reduced due to pseudo feature points

Engineering Contradiction:
Improveease of image acquisitionVSAvoidmatching accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent changes the parameters used for feature point evaluation by introducing direction difference calculations and geometric relationship verification. Instead of relying solely on image quality or simple feature detection, the system evaluates feature points based on their directional consistency with surrounding points and their geometric relationships, making the matching process robust to variations in image quality and preprocessing strength

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11710342B2Fingerprint matching method and apparatus, electronic equipment and readable storage medium
Publication Date: 2023.07.25 SHANGHAI HARVEST INTELLIGENCE TECH CO LTD
  • US11710342B2 patent drawing
  • US11710342B2 patent drawing
  • US11710342B2 patent drawing

AI summary

The present disclosure provides a fingerprint matching method and apparatus, an electronic equipment and a readable storage medium. The method includes: extracting a plurality of to-be-matched feature points from the to-be-identified fingerprint image; performing a first matching between the plurality of to-be-matched feature points and a plurality of template feature points in the template fingerprint image, wherein the first matching includes: identifying true feature points in the plurality of to-be-matched feature points, and determining feature point pairs each of which includes a true feature point and a template feature point corresponding to the true feature point in the template fingerprint image as a first matching result; removing at least one falsely matched feature point pair from the first matching result; and performing a second matching between the to-be-identified fingerprint image and the template fingerprint image based on remaining feature point pairs in the first matching result.