Facial Authentication Thresholding for Biometric Aging Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition systems face challenges in handling biometric temporal changes, such as aging and changes in appearance, which affect confidence values, making them unreliable as a sole means of access control, especially in ultra-secure environments, and there are concerns about the wide distribution and storage of biometric data.
Innovation Solution
Facial recognition is integrated with a user's credential device, allowing real-time template generation and comparison directly on the device, combined with additional authentication factors to dynamically adjust confidence thresholds based on available modalities, ensuring secure access control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If facial recognition is used as a sole authentication factor, then user convenience is improved, but reliability deteriorates due to biometric temporal changes
Solution Approach 1:
The system dynamically adjusts authentication requirements based on confidence values. When biometric confidence is high, the system accepts facial recognition alone. When confidence drops due to temporal changes, the system automatically requests additional authentication factors, creating a dynamic authentication process that adapts to current conditions.
Solution Approach 2:
The system changes the authentication threshold parameter based on confidence value analysis. Instead of using a fixed confidence threshold, the system adjusts the required confidence level and authentication factor requirements based on the detected confidence value, which varies with biometric temporal changes.
2Adaptability or versatility
If confidence thresholds are lowered to accommodate biometric changes, then adaptability is improved, but measurement precision deteriorates
Solution Approach 1:
The confidence threshold is not fixed but dynamically adjusted based on the detected confidence value. The system monitors confidence values over time and adapts the threshold requirements accordingly, allowing higher thresholds when confidence is high and lower thresholds when confidence decreases due to temporal changes.
Solution Approach 2:
The system performs preliminary analysis of confidence values before final authentication decisions. By evaluating confidence trends and patterns in advance, the system can prepare appropriate threshold adjustments and authentication factor requirements, preventing false rejections while maintaining security.
3Reliability
If multiple authentication factors are required, then reliability is improved, but device complexity increases
Solution Approach 1:
The system dynamically determines the number and type of authentication factors required based on real-time confidence value assessment. Instead of always requiring multiple factors, the system adapts the authentication process to require only what is necessary at that moment, reducing complexity when possible while maintaining reliability when needed.
Solution Approach 2:
The authentication process is segmented into optional stages. The system first attempts facial recognition alone, and only if confidence is insufficient does it proceed to request additional authentication factors. This segmentation allows the system to maintain simplicity for most cases while having the capability to increase security when needed.
4Adaptability or versatility
If biometric data is widely distributed for recognition, then adaptability is improved, but security worsens due to data breach risks
Solution Approach 1:
The system changes the parameter of data distribution from widespread to localized. Instead of distributing biometric templates across multiple devices or servers, the system performs facial recognition processing locally at the access device, requiring only transmission of encrypted biometric data for authentication decisions.
Solution Approach 2:
The system introduces an intermediary processing step where biometric data is transmitted in encrypted form and processed through secure channels. The actual facial recognition and confidence value calculation occur at the access device, minimizing the need for widespread biometric data storage and reducing security risks associated with data distribution.
Data Source
AI summary
A computer readable medium having executable code to: receive at least one of a first image of the user or a first representation of a face of the user; if a first image of the user was received, then generate a generated representation of the face of the user using the first image; capture a second image of the user and generate a second representation of the face of the user using the second image; receiving an authentication factor; determine validity of the authentication factor; determine a confidence of a match between the second representation and at least one of the first representation and the generated representation; and if the confidence is below a threshold and the authentication factor is determined to be valid, at least one of supplement or replace at least one of the first image or first representation with the second image or second representation, respectively.