Biometric Authentication Search Space Reduction via Kiosk Identifier
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in efficiently authenticating users in high-traffic contexts, such as quick service restaurants, due to the need to compare captured images with a large number of reference images, leading to increased processing time and potential false positives.
Innovation Solution
A biometric authentication system deployed in kiosk devices that uses a transceiver to broadcast a kiosk identifier, captures user images with a camera, and compares them to retrieved reference images from a server system, generating a score value for authentication, which reduces the search space and processing time by identifying unique identifiers before user engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If biometric authentication systems compare captured images with a large number of reference images, then authentication coverage increases, but processing time increases and accuracy decreases
Solution Approach 1:
The system performs preliminary actions by receiving unique identifiers from mobile devices before the user actually engages with the kiosk. This allows the system to pre-fetch and prepare reference images in advance, so that when authentication is needed, the comparison can be performed much faster against a pre-loaded subset rather than searching through all reference images from scratch.
Solution Approach 2:
The system segments the large database of reference images into smaller, manageable subsets based on unique identifiers received from mobile devices. Instead of comparing against all reference images simultaneously, the system divides the search space into targeted groups, making the comparison process more efficient and reducing processing time while maintaining comprehensive coverage.
2Adaptability or versatility
If biometric authentication systems compare captured images with a large number of reference images, then authentication coverage increases, but measurement precision decreases
Solution Approach 1:
By receiving unique identifiers and preparing reference images in advance before user engagement, the system ensures that only relevant reference images are loaded into memory for comparison. This preliminary preparation reduces memory access time and ensures that the comparison is performed against the most relevant data, improving both speed and accuracy.
Solution Approach 2:
The system applies local quality by focusing computational resources on comparing the captured image with a specific subset of reference images identified through unique identifiers, rather than uniformly processing all reference images. This targeted approach improves measurement precision by concentrating analysis on the most relevant comparisons.
3Speed
If biometric authentication systems process images in real-time at the kiosk, then response time is fast, but system complexity increases
Solution Approach 1:
The system extracts the unique identifier reception and reference image preparation functions from the main authentication processing flow. By separating these preliminary steps from the core image comparison operation, the system reduces the complexity of the real-time processing required at the kiosk while maintaining fast response times through pre-computed preparations.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for a biometric authentication system. In one aspect, a method includes broadcasting, through a transceiver, a signal that includes a kiosk identifier; receiving, from a server system, a reference image of the user based on a mobile computing device associated with the user having received the signal and providing the kiosk identifier to the server system; capturing, with a camera, an image of the user; and authenticating the user based on a comparison of the captured image of the user to the reference image of the user based on a comparison of the captured image of the user to the reference image of the user.