Facial Recognition Threshold Adjustment for Feature Distinctiveness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition systems fail to authenticate users accurately due to removable and non-removable facial features that decrease the distinctiveness between faces, leading to unauthorized access or denial of authorized access.
Innovation Solution
A method that captures and analyzes facial images to detect removable features like sunglasses and non-removable features like facial hair, prompting users to remove the former and adjusting similarity score thresholds for the latter, ensuring accurate authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If facial recognition systems use standard similarity score thresholds for authentication, then the system operates with fixed security levels, but it cannot accurately distinguish between authorized and unauthorized users when removable or non-removable facial features are present
Solution Approach 1:
The system dynamically adjusts the similarity score threshold based on detected facial features. When removable features (sunglasses, hats) or non-removable features (facial hair, scars) are detected, the threshold is modified to account for the reduced distinctiveness, allowing the system to adapt to varying facial conditions while maintaining security
Solution Approach 2:
The system changes the parameter of similarity score threshold based on the presence of specific facial features. By detecting features that decrease distinctiveness and adjusting the threshold accordingly, the system maintains accurate authentication across different facial conditions without requiring multiple fixed threshold levels
2Reliability
If the system does not account for facial features that decrease distinctiveness, then authentication is faster and simpler, but erroneous authentication attempts increase and authorized access is denied
Solution Approach 1:
The system performs preliminary detection of facial features that decrease distinctiveness before the main authentication comparison. By identifying removable and non-removable features in advance and adjusting the similarity threshold accordingly, the system prevents erroneous authentications and maintains reliability without adding complex multi-stage verification processes
3Measurement precision
If the system prompts users to remove removable facial features, then authentication accuracy improves, but user convenience and access speed decrease
Solution Approach 1:
The system provides feedback to users when removable or non-removable facial features are detected. Users are notified of the specific feature detected and given guidance on how to improve authentication (e.g., removing sunglasses or covering facial hair), enabling them to correct the issue and complete authentication successfully
Data Source
AI summary
An example method includes capturing, by a camera of a computing device, an image including at least a face of a user, calculating a face template of the face of the user in the image, and analyzing the face template to determine whether the face includes at least one of a removable facial feature that decreases a level of distinctiveness between two faces and a non-removable facial feature that decreases a level of distinctiveness between two faces. When the face includes the removable facial feature, the method further includes outputting a notification for the user to remove the removable facial feature. When the face includes the non-removable facial feature, the method further includes adjusting a first similarity score threshold to a second similarity score threshold.