Face Authentication Using Age-Corrected Similarity Matching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing authentication systems struggle to accurately identify individuals whose facial appearance has changed over time, leading to recognition errors.

Innovation Solution

An authentication system that acquires first and second images of an individual at different timings, along with age information and a time period, calculates similarity, and corrects it based on age and time elapsed to determine identity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If facial image comparison is performed without age correction, then the authentication process is simple and fast, but recognition accuracy decreases for individuals whose appearance has changed over time

Engineering Contradiction:
Improverecognition accuracyVSAvoidauthentication process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by acquiring age information and time period data before conducting the facial image comparison. These preliminary data acquisitions enable the subsequent correction of similarity scores based on aging effects, thereby improving recognition accuracy for individuals whose appearance has changed over time without significantly complicating the overall authentication process

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes the parameter being measured by introducing age information and time period as correction factors. Instead of directly comparing facial images, the system calculates a similarity score and then adjusts this score based on aging-related parameters, transforming a simple image comparison into an age-compensated authentication process that maintains accuracy while controlling complexity

Inventive Principle:
Principle #35Parameter changes

2Reliability

If multiple parameters (age information, time period) are acquired and processed, then authentication accuracy improves, but processing time and computational load increase

Engineering Contradiction:
Improveauthentication reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Age information and time period data are acquired as preliminary information before the main authentication process. By preparing these correction parameters in advance, the system can efficiently adjust similarity scores without requiring complex real-time calculations during the authentication decision phase, thereby reducing overall processing time while maintaining high authentication reliability

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system replaces complex mechanical or manual authentication processes with automated computational methods. By using algorithms to calculate similarity scores and apply age-based corrections automatically, the system reduces processing time and computational load compared to manual verification methods, while significantly improving authentication reliability through consistent application of correction factors

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3690806B1Authentication device, authentication system, authentication method, and program
Publication Date: 2026.02.25 PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
  • EP3690806B1 patent drawingFigure 1
  • EP3690806B1 patent drawingFigure 2
  • EP3690806B1 patent drawingFigure 3

AI summary

An authentication device includes an acquisition unit that acquires first image data generated at a first timing and indicating a first face of a living creature to be authenticated, age information indicating an age of the living creature at the first timing, second image data generated at a second timing later than the first timing and indicating a second face of the living creature, and aging information indicating a time period from the first timing to the second timing, and a controller that compares the first image data with the second image data. The controller calculates a similarity between the first face and the second face based on the first image data and the second image data, corrects the similarity based on the age information and the aging information, and determines, based on the corrected similarity, whether the living creature in the first image data is identical to the living creature in the second image data.