Biometric Authentication Device Surface Reflection Removal

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Biometric authentication techniques face challenges in accurately acquiring biological information due to surface reflection effects, which can obscure or distort images of fingerprints, veins, and faces, leading to difficulties in obtaining reliable authentication results.

Innovation Solution

A biometric authentication device that repeatedly images a biological part while changing its relative position, detects pixels with surface reflection components based on temporal variation, and corrects pixel values to generate images with no surface reflection, enabling accurate authentication.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If surface reflection is removed using polarization filters or multiple illumination sources, then measurement precision of biological information is improved, but device complexity increases

Engineering Contradiction:
Improveprecision of biological information acquisitionVSAvoidcomplexity of optical system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and removes the surface reflection component from the captured image through image processing algorithms. By separating the surface reflection component from the diffuse reflection component containing the biological information, the method eliminates the need for complex optical systems while improving measurement precision of vein patterns and other biological features.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the mechanical/optical approach (polarization filters, multiple illumination sources) with a computational approach (image processing algorithms). The surface reflection removal is achieved through software-based component separation rather than hardware-based optical manipulation, thereby reducing device complexity while maintaining measurement precision.

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

2Ease of operation

If surface reflection components are included in the captured image, then ease of operation is improved (simpler imaging process), but measurement precision deteriorates due to obscured biological information

Engineering Contradiction:
Improvesimplicity of imaging processVSAvoidaccuracy of biological information
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent converts the harmful surface reflection component into a useful element by explicitly modeling and separating it from the image. Instead of treating surface reflection as noise to be eliminated through complex optics, the method uses image processing to identify and remove only the surface reflection component while preserving the diffuse reflection component containing the biological information, thereby maintaining ease of operation while improving measurement precision.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

3Measurement precision

If multiple images are captured from different angles to remove surface reflection, then measurement precision is improved, but loss of time increases

Engineering Contradiction:
Improveaccuracy of biological informationVSAvoidtime for image acquisition
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by capturing a single image that contains both surface reflection and diffuse reflection components, then uses image processing algorithms to separate these components. This approach eliminates the need to capture multiple images from different angles, thereby reducing time loss while maintaining measurement precision through computational separation of reflection components.

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This method effectively removes surface reflection effects, improving the accuracy of biometric authentication without the need for complex optical systems or special filters, thus enhancing the reliability and efficiency of the authentication process.

Implementation Method 1

an image of a biological part such as a fingerprint, a vein, and a face is first acquired preliminarily as biological information of a user

Methodology Applied
Scientific EffectReflection: Reflection

Implementation Method 2

Surface reflection is a phenomenon that light reflects on a surface of an object. When a surface of an object is planar, light regularly reflects due to surface reflection

Methodology Applied
Scientific EffectSurface reflection: Reflection

Implementation Method 3

Diffuse reflection is a phenomenon that incident light penetrates the inside of an object, repeats diffusion, and then, is radiated from a surface of the object

Methodology Applied
Scientific EffectDiffuse reflection: Scattering

Data Source

PatentEP2562687B1Biometric authentication device and method
Publication Date: 2020.01.15 FUJITSU LTD
  • EP2562687B1 patent drawingFigure 1A~1B
  • EP2562687B1 patent drawingFigure 2
  • EP2562687B1 patent drawingFigure 3A~3B

AI summary

A biometric authentication device includes, an acquisition unit configured to repeatedly image a biological part of an authenticated person while changing a relative position with respect to the biological part so as to acquire time-series biological images; a detection unit configured to detect a pixel corresponding among the biological images from the time-series biological images; an extraction unit configured to extract a pixel of which a pixel value includes a surface reflection component from the biological part, from each of the time-series biological images, on the basis of a degree of divergence of temporal variation of a pixel value of the detected pixel in the time-series biological images from an estimation result of temporal variation; a generation unit configured to correct the pixel value of the extracted pixel on the basis of a pixel value; and an authentication unit configured to perform personal authentication of the authenticated person.