Face Authentication Switching Detection Using Contour and Reflection Cues

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing face authentication systems struggle to detect switching of imaging targets during authentication, allowing unauthorized access through the use of photographs, which can bypass action detection and compromise security.

Innovation Solution

A face authentication method that calculates contour and reflection change degrees to detect switching of imaging targets by comparing consecutive images, using edge statistics and pixel value statistics to determine if the imaging target has changed, and incorporates action detection to verify the authenticity of the user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If action detection is performed to prevent photograph impersonation, then authentication security is improved, but the system cannot detect switching of imaging targets during authentication

Engineering Contradiction:
Improveauthentication securityVSAvoidswitching detection capability
Core Design Contradiction:
ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The system performs preliminary analysis of reflection characteristics and contour changes in images captured during authentication. By pre-calculating reflection change degrees and contour change degrees for consecutive images, the system can detect switching of imaging targets before final authentication decision is made, preventing photograph impersonation attempts.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors image characteristics during authentication and provides feedback by comparing reflection characteristics and contour changes between consecutive images. When switching of imaging targets is detected through this feedback mechanism, the authentication process is terminated, ensuring security against photograph impersonation.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If reflection characteristics are used to detect imaging target switching, then switching detection accuracy is improved, but image blur reduces detection reliability

Engineering Contradiction:
Improveswitching detection accuracyVSAvoiddetection reliability under image blur
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The system segments the detection task into two independent components: reflection characteristic analysis and contour change analysis. By separating these detection functions, the system can use reflection characteristics for fine-grained detection while relying on contour changes for robust detection under blur conditions, ensuring overall reliability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes the detection parameters dynamically based on image quality. When image blur is detected, the system increases reliance on contour change degree for switching detection, while using reflection change degree for clearer images. This adaptive parameter adjustment maintains detection reliability across varying image quality conditions.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If multiple images are captured for action detection, then authentication accuracy is improved, but processing time increases

Engineering Contradiction:
Improveauthentication accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary switching detection using reflection and contour change analysis on images captured during the authentication process. By conducting this preliminary analysis in parallel with the main authentication flow, the system can quickly identify and reject photograph impersonation attempts without significantly increasing overall processing time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system extracts key detection features (reflection characteristics and contour changes) from the sequence of captured images, focusing computational resources only on these critical features rather than processing all image data. This extraction approach maintains high authentication accuracy while minimizing processing time.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP4216149B1Face authentication method, face authentication program, and face authentication device
Publication Date: 2026.02.11 FUJITSU LTD
  • EP4216149B1 patent drawingFigure 1
  • EP4216149B1 patent drawingFigure 2
  • EP4216149B1 patent drawingFigure 3

AI summary

It is possible to easily detect switching of imaging targets in the middle of authentication. A face authentication device (10) acquires a plurality of time-series images (4a, 4b, 4c,...) from a camera (3). Next, the face authentication device (10) calculates a contour change degree representing a difference between contours included in face images respectively imaged in a first image (5) and a second image (6 and 7) that are consecutive in chronological order. Furthermore, the face authentication device (10) calculates a reflection change degree representing a difference between light reflection characteristics of the first image (5) and the second image (6 and 7). Then, the face authentication device (10) determines whether or not the switching of the imaging targets of the camera (3) has occurred, based on the contour change degree and the reflection change degree.