Fingerprint Authentication Using Local and Global Image Correction

Solution Overview

Problem

Conventional authentication methods fail to achieve optimal correction of image data for biometric authentication due to limitations in noise handling and detailed information maintenance, with local correction being unsuitable for high-noise images and global correction being inadequate for high-quality images.

Innovation Solution

An authentication apparatus that separately performs local and global corrections on fingerprint images, calculates the difference between the two, and selects the appropriate correction method based on a threshold to ensure suitability for the image quality, using a Sobel filter and smoothing filter for detailed information maintenance and noise removal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If local correction is used, then detailed line information is maintained, but noise larger than the smoothing filter cannot be corrected

Engineering Contradiction:

Improvedetailed line informationVSAvoidnoise

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The correction process is segmented into two independent correction units: local correction unit that processes detailed line information and global correction unit that handles noise. Each unit operates separately on different aspects of the image data, allowing both detailed information preservation and noise removal without interference between the two functions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system changes the correction parameter by selecting between local and global correction methods based on the image quality assessment. When noise is detected, global correction parameters are applied; when detailed information is prioritized, local correction parameters are used. This dynamic parameter adjustment allows the system to adapt to different image conditions.

Inventive Principle:

Principle #35Parameter changes

2Object-affected harmful factors

If global correction is used, then noise is removed, but detailed line information cannot be maintained

Engineering Contradiction:

ImprovenoiseVSAvoiddetailed line information

Core Design Contradiction:

Object-affected harmful factorsVSMeasurement precision

Solution Approach 1:

The correction process is segmented into two independent correction units: local correction unit that processes detailed line information and global correction unit that handles noise. Each unit operates separately on different aspects of the image data, allowing both detailed information preservation and noise removal without interference between the two functions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system changes the correction parameter by selecting between local and global correction methods based on the image quality assessment. When noise is detected, global correction parameters are applied; when detailed information is prioritized, local correction parameters are used. This dynamic parameter adjustment allows the system to adapt to different image conditions.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If a single correction method is used, then the process is simple, but optimum correction cannot be obtained for all image qualities

Engineering Contradiction:

Improvecorrection processVSAvoidauthentication precision

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The correction system is made dynamic by introducing an image quality assessment unit that evaluates the input image and automatically selects the appropriate correction method. The system transitions from a static single-method approach to a dynamic multi-method approach where the correction strategy adapts based on real-time image quality analysis, optimizing authentication precision for different image conditions.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary image quality assessment before applying correction to determine which correction method is most suitable. This preliminary action allows the system to prepare the appropriate correction parameters in advance, ensuring that the optimal correction method is applied without adding complexity during the actual correction process.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentEP2296113B1Authentication device, authentication method and authentication program

Publication Date: 2019.11.13 FUJITSU LTD

Figure 1
Figure 2
Figure 3~4

AI summary

An authentication apparatus performs local and global corrections on image data. Local correction uses the shape of a local line indicated by line information contained in a neighboring region on image data whose feature value extracted from the shape of a line is used for authentication. Global correction uses the shape of a global line indicated by line information contained in a region larger than the neighboring region. The authentication apparatus calculates the difference between line information contained in image data corrected by local correction and line information contained in image data corrected by global correction and compares the difference with a threshold. If the difference is less than the threshold, the authentication apparatus outputs, as line information contained in image data, line information corrected by local correction and, if the difference is greater, it outputs, as line information contained in image data, line information corrected by global correction.