Face Authentication System with Segmented Matching Servers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing face authentication systems lack a configuration that allows for stable and efficient operation for individual providers of face authentication services, as they typically require both face feature and image data to be stored in a cloud server, which is not scalable or secure.
Innovation Solution
A system comprising multiple face authentication machines, a server device, and a terminal device, where the server stores and manages face image data and generates face feature data for matching, allowing each authentication machine to request matching operations from identified face image matchers based on association information, reducing computational load and eliminating the need for data storage on authentication machines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If face feature amount data and face image data are stored in a cloud server for centralized management, then system stability and data security are improved, but device complexity and operational efficiency deteriorate due to the need for multiple storage locations and data transmission channels
Solution Approach 1:
The patent divides the authentication system into two functional segments: face authentication machines that perform image capture and feature extraction, and face image matchers that store and compare face image data. This segmentation allows each component to have specialized functions, reducing overall system complexity while maintaining centralized secure storage of sensitive face image data on the matcher side.
2Speed
If face image data is stored on face authentication machines for local processing, then processing speed is improved, but data security and loss prevention worsen due to potential theft or attacks on authentication machines
Solution Approach 1:
The patent extracts the sensitive face image data storage function from the face authentication machines and places it exclusively in the face image matchers. The authentication machines only handle non-sensitive face feature amount data and image capture, while the matchers securely store and manage the actual face image data, preventing data exposure even if authentication machines are compromised.
3Device complexity
If a centralized cloud server stores all face data for all providers, then data management is simplified, but scalability and individual provider autonomy deteriorate
Solution Approach 1:
The patent introduces face image matchers as intermediary components that each provider can deploy independently. These matchers act as local intermediaries between the authentication machines and the centralized cloud infrastructure, allowing each provider to maintain autonomous control over their data while still benefiting from standardized protocols and potential cloud connectivity when needed.
Data Source
AI summary
Provided is a system for face authentication which can operate separately for individual providers of face authentication service in a stable and efficient manner. The system includes: face authentication machines; an administrator terminal; a face management server for storing face image data of registered users; and face matching servers, each configured to generate face feature amount data of a person to be verified from image data acquired from the camera of a face authentication machine and to perform a matching operation by comparing the face feature amount data of the person with that of registered users. Prior to face authentication, data of associations between face authentication machines and face matching servers entered by an administrator is transmitted from the administrator terminal to the face management server.


