Liveness Detection Module Using Optical Interference Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems are susceptible to spoofing attacks, particularly due to the modality-specific nature of presentation attacks, which existing liveness detection methods struggle to effectively counteract, especially when high-quality biometric data is used for verification.
Innovation Solution
An image capturing system with a laser unit emitting over 20 mW of power, utilizing a conductive layer with an inclined surface to reflect light, combined with a data processing module that generates liveness detection signals from multiple sequential image signals, enhancing signal-to-noise ratio and reducing ambient light interference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If liveness detection is performed using existing image processing techniques, then the detection can be implemented, but the user experience is degraded due to unexpected interruptions and the security is insufficient against sophisticated spoofing attacks
Solution Approach 1:
The system performs preliminary liveness detection by analyzing user interactions with virtual objects before the actual authentication process. This preliminary action includes detecting micro-movements, analyzing interaction patterns, and verifying liveness in advance, which prevents sophisticated spoofing attacks while maintaining a natural user experience without unexpected interruptions
Solution Approach 2:
The system introduces virtual objects as an intermediary between the user and the liveness detection process. These virtual objects serve as a mediator that enables natural interaction-based liveness verification, allowing the system to detect genuine human presence through interaction patterns while maintaining user experience quality
2Reliability
If traditional liveness detection methods are used, then the implementation is straightforward, but the security against sophisticated spoofing attacks is insufficient
Solution Approach 1:
The system transitions from static liveness detection to dynamic interaction-based detection. By analyzing real-time user interactions with virtual objects, including micro-movements and interaction patterns, the system achieves superior security against spoofing attacks while using existing image processing techniques, thereby managing complexity effectively
Solution Approach 2:
The system changes the detection parameters from traditional static features to dynamic interaction patterns. By monitoring how users interact with virtual objects over time, the system extracts liveness indicators that are difficult to spoof, enhancing security without requiring fundamentally new detection mechanisms
Data Source
Figure 1A~1B
Figure 2
Figure 3~4
AI summary
The present invention provides a liveness detection apparatus and a related authentication method. The liveness detection apparatus comprises a light emitting unit, an image sensing unit, and a data processing module. A substrate of the light emitting unit has a first inclined surface, and emitted light is reflected light of the first inclined surface. The present invention triggers an authentication process by means of an application to authenticate a user. The light emitting unit emits light of a specific pattern in a specific period, to produce a plurality of image signals. The present invention produces a plurality of liveness detection signals for determining a living body, wherein the present invention calculates each optical interference pattern from at least one sequential image signal to produce the plurality of liveness detection signals. When a liveness score of the plurality of liveness detection signals reaches a liveness threshold, the present invention further produces feature recognition data for matching, wherein the feature recognition data is produced by calculating each optical interference pattern from at least one sequential image signal. The present invention then compares the feature recognition data with previously registered data to lock or unlock the liveness detection apparatus and/or a system coupled thereto.