Biometric Identification Using Segmented Fingerprint Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The use of classification into classes for 1:N matching in biometric authentication can lead to reduced authentication accuracy due to insufficient information, resulting in high false reject and false accept rates, especially when the sensor device is smaller than the user's body part, causing failures in classification during registration or identification.
Innovation Solution
An identification apparatus that classifies input biometric data into two or more classes based on features, calculates similarity with registered data, and identifies the user by matching against these classes, using a classification unit, calculation unit, and identification unit, with features like ridge endings and bifurcations extracted from fingerprint images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If classification into classes is used for 1:N matching, then matching efficiency is improved, but authentication accuracy deteriorates due to insufficient information from small sensor devices
Solution Approach 1:
The patent segments the fingerprint image into multiple regions (first region containing ridge bifurcations, second region containing ridge endings) and extracts feature information from each region separately. This segmentation allows the system to maximize the use of limited sensor data by focusing on the most informative areas, thereby maintaining authentication accuracy while improving matching efficiency through region-based classification.
2Area of stationary object
If the sensor device size is reduced, then device compactness is improved, but classification reliability deteriorates due to insufficient information
Solution Approach 1:
The patent applies local quality by focusing feature extraction on specific high-value regions within the limited sensor area. The first region captures ridge bifurcations and the second region captures ridge endings, which are the most discriminative features. This localized approach maximizes the information quality from the constrained sensor device area, maintaining classification reliability despite the small sensor size.
3Speed
If classification into classes is performed, then matching speed is improved, but false reject rate increases when classification fails
Solution Approach 1:
The patent performs preliminary feature extraction and classification during the registration phase, pre-computing the class assignments for stored fingerprint images. This preliminary action enables fast matching during authentication by simply comparing the class of the input fingerprint with pre-stored classes, achieving high matching speed while minimizing false rejects through accurate preliminary classification.
4Reliability
If all registered biometric data is matched against entered data, then authentication thoroughness is improved, but processing time increases
Solution Approach 1:
The patent segments the authentication process into two stages: first, classifying the input biometric data into predetermined classes; second, matching only against registered data within the same class. This segmentation dramatically reduces the number of comparisons needed while maintaining thoroughness, as all data within the relevant class is still examined, achieving both authentication thoroughness and reduced processing time.
Data Source
AI summary
An identification apparatus includes a classification unit that determines two or more classes into which input biometric data is classified out of a plurality of classes based on features of the input biometric data, where a plurality of items of registered biometric data have been classified into at least one of the plurality of classes, a calculation unit that calculates similarity between the input biometric data and each item of the registered biometric data registered in each of the two or more classes into which the input biometric data is classified, and an identification unit that identifies data on a user who has entered the input biometric data among the registered biometric data registered in any of the two or more classes into which the input biometric data is classified, based on the similarity to the input biometric data.