Authentication Device Reliability Control via Image Quality Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face authentication methods fail to prevent erroneous authentication when images with low reliability, such as those with blur, noise, or incorrect detection, are used, leading to potential misidentification of individuals.
Innovation Solution
An authentication device that generates feature amounts from images, acquires registered data including feature amounts and associated unsuitable authentication information, calculates similarity between generated and registered feature amounts, and performs authentication only when similarity meets a predetermined condition and the unsuitable authentication flag is false.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If face authentication is performed using extracted feature amounts, then authentication functionality is achieved, but erroneous authentication occurs when low-quality images (blur, noise, incorrect detection) are used
Solution Approach 1:
The system performs preliminary quality assessment of the input image by detecting reliability indicators (blur, noise, detection accuracy) before extracting feature amounts. Only images meeting predetermined quality thresholds proceed to authentication processing, preventing low-quality images from causing erroneous authentication
Solution Approach 2:
A reliability determination unit acts as an intermediary between image input and authentication processing. This intermediary evaluates image quality and controls the flow of processing, blocking low-quality images from entering the feature extraction and authentication stages
2Reliability
If a reliability threshold is set for authentication, then some erroneous authentication is prevented, but images slightly exceeding the threshold still cause erroneous authentication
Solution Approach 1:
The authentication process is segmented into distinct stages: reliability evaluation, quality threshold checking, and authentication processing. Each stage handles specific aspects of quality control, allowing for more granular and effective prevention of erroneous authentication compared to a single threshold check
3Reliability
If multiple quality checks are performed before authentication, then erroneous authentication is reduced, but processing time increases
Solution Approach 1:
Quality checks are performed as preliminary actions before main authentication processing. By filtering out low-quality images early through rapid reliability indicators (blur detection, noise detection, detection confidence), the system avoids wasting processing time on images that would fail authentication anyway
Data Source
AI summary
An authentication device includes a feature amount generation unit for generating a feature amount of an object in an image, a registered data acquisition unit for acquiring registered data where a feature amount of a predetermined object in an image is registered in advance as a registered feature amount, and NG information associated with the registered feature amount unsuitable for authentication is recorded, a similarity acquisition unit for acquiring a similarity between the predetermined feature amount and the registered feature amount acquired from the registered data acquisition unit, and a determination unit for performing authentication if the degree of similarity acquired by the degree of similarity acquisition unit satisfies a predetermined authentication condition, and even if the degree of similarity satisfies the predetermined authentication condition, if the NG information associated with the registered feature amount is acquired, the predetermined feature amount is not authenticated.