SST Facial Authentication with Background Pixel Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric facial authentication systems, particularly in Self-Service Terminals (SSTs), are vulnerable to spoofing using photographs or videos, leading to potential monetary losses for financial institutions and users.
Innovation Solution
A method and system for SST facial authentication that samples and authenticates both user face and background pixels using machine learning algorithms to differentiate between live and non-live images by analyzing pixel values and light attributes, continually training on environmental conditions to prevent unauthorized access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional facial authentication is used, then authentication speed is fast, but security is vulnerable to spoofing
Solution Approach 1:
The authentication process is segmented into multiple independent verification stages: face detection, facial feature extraction, background region detection, and depth analysis. Each stage operates independently and must all pass for authentication to succeed, making the system more secure without requiring a single complex authentication mechanism
Solution Approach 2:
The system transitions from two-dimensional image analysis to three-dimensional depth analysis by capturing depth information alongside facial images. This adds a new dimension of verification that effectively prevents spoofing with photographs or videos, as these cannot provide accurate depth data
2Reliability
If background pixel sampling is added to authentication, then spoofing resistance improves, but processing time increases
Solution Approach 1:
Background pixel sampling and environmental condition analysis are performed concurrently with facial feature extraction rather than sequentially. The system prepares background region data and depth information in advance during the same processing cycle, reducing overall authentication time while maintaining comprehensive verification
Solution Approach 2:
The system samples pixels from specific critical background regions rather than analyzing the entire background image. By focusing on key areas where spoofing attempts are most likely to fail, the system achieves effective anti-spoofing capability with reduced processing overhead
3Measurement precision
If depth analysis is performed, then authentication accuracy improves, but system complexity increases
Solution Approach 1:
The depth analysis component serves multiple functions simultaneously: it verifies three-dimensional presence of the user, detects spoofing attempts with photographs or videos, and provides additional authentication data for the facial recognition algorithm. This multi-functionality justifies the added system complexity through enhanced security benefits
Data Source
AI summary
Real-time facial recognition is augmented with a machine-learning process that samples pixels from images captured for the physical environmental background of a device, which captures an image of a user's face for facial authentication. The background pixel points that are present in a captured image of a user's face from a camera of the device are authenticated with the image of the user's face. The value of the background pixel points are compared against the expected values for the background pixel points provided by the on-going machine-learning process for the background.