Biometric Authentication Quality Estimation via User Posture

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing biometric authentication systems face challenges in obtaining suitable fingerprint images for authentication due to variations in user posture, despite guidance images being displayed.

Innovation Solution

An information processing apparatus that includes a posture information acquisition unit, a learned model trained through machine learning to estimate the quality of biometric information based on user posture, and an output unit that provides estimated quality information to adjust the user's posture for improved authentication.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If guidance image is displayed to guide user's finger position, then ease of operation is improved, but reliability of authentication is worsened because the imaging unit still cannot obtain suitable fingerprint images depending on user posture

Engineering Contradiction:

Improveguidance image displayVSAvoidauthentication reliability

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system displays a quality indicator showing the estimated quality of the fingerprint image based on detected posture information, providing feedback to the user about whether their current posture will yield a suitable image for authentication

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system estimates fingerprint image quality based on posture information before the imaging unit actually captures the image, allowing the user to adjust their posture in advance to avoid poor quality images

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If machine learning model is introduced to estimate quality from posture information, then measurement precision is improved, but device complexity is worsened

Engineering Contradiction:

Improvequality estimation precisionVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system introduces a quality indicator as an intermediary element that conveys the estimated quality information from the machine learning model to the user in a simple, understandable format, bridging the gap between complex processing and simple interaction

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250200964A1Information processing apparatus, authentication system, information processing method, non-transitory computer-readable medium, learned model, and method for generating learned model

Publication Date: 2025.06.19 NEC CORP

AI summary

Provided is an information processing apparatus capable of estimating quality of biometric information according to a posture of a user. An information processing apparatus (100) according to the disclosure includes: a posture information acquisition unit (101) that acquires posture information indicating a posture of a user who performs biometric authentication; a learned model (102) learned to estimate quality information from input posture information and output estimated quality information by performing machine learning using the posture information and quality information indicating quality of biometric information corresponding to the posture information as teacher data; and an output unit (103) that outputs the estimated quality information from the acquired posture information using the learned model (102).