Biometric Liveness Detection via Motion Vector Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current live-ness detection techniques are ineffective in distinguishing between genuine and fraudulent biometric data during remote authentication transactions, as imposters can easily spoof biometric data using obtained images or eavesdropped information, making it difficult to verify the physical presence of the user.
Innovation Solution
A method and device that generate a motion type feature vector and liveness rating feature vector from captured biometric data frames, merge these vectors to predict spoofing, and adjust the user head motion type prediction to 'no motion' when spoofing is detected, using a processor and memory to store and calculate predictions for accurate authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional biometric authentication is used, then authentication speed is fast, but authentication reliability is low due to spoofing attacks
Solution Approach 1:
The authentication system is segmented into multiple independent components: traditional biometric verification and liveness detection. The liveness detection module independently analyzes motion patterns, blinking behavior, and physiological signals, then combines results with biometric verification to achieve reliable authentication without requiring complete system redesign
Solution Approach 2:
A liveness detection intermediary module is introduced between the biometric verification system and the authentication decision. This intermediary analyzes motion patterns and physiological indicators to determine authenticity, preventing spoofed biometric data from compromising the authentication system while maintaining the original biometric verification capability
2Reliability
If liveness detection is added to detect spoofing, then authentication reliability improves, but processing time increases
Solution Approach 1:
Liveness detection analysis is performed preliminarily during the biometric data capture phase rather than as a separate post-processing step. Motion patterns, blinking behavior, and physiological signals are analyzed concurrently with biometric feature extraction, allowing authentication reliability to improve without significant time penalty
Solution Approach 2:
The liveness detection process maintains continuous operation throughout the authentication transaction by analyzing motion patterns and physiological indicators in real-time as biometric data is captured. This continuous analysis ensures reliable spoofing detection while minimizing additional processing time through efficient parallel computation
3Measurement precision
If motion analysis is performed on all users, then spoofing detection accuracy improves, but computational load increases
Solution Approach 1:
Motion analysis is applied selectively rather than uniformly to all authentication transactions. The system performs basic motion pattern analysis on all users and applies more intensive computational liveness detection only when anomalies are detected or risk levels warrant additional verification, improving spoofing detection accuracy while reducing overall computational energy consumption
Solution Approach 2:
The system dynamically adjusts motion analysis parameters based on authentication risk levels and user behavior patterns. Computational resources are allocated adaptively, with motion detection sensitivity and analysis depth modified in real-time to balance spoofing detection accuracy against energy consumption, preventing unnecessary computational overhead for low-risk transactions
Data Source
AI summary
A method for detecting user head motion during an authentication transaction is provided that includes generating, by a processor, a motion type feature vector and a user head motion type prediction based on data generated for a sequence of frames. The frames are included in biometric data captured from a user. Moreover, the method includes generating a liveness rating feature vector based on the generated frame data, merging the motion type and liveness rating vectors, and generating a spoof prediction from the merged vector. When the generated spoof prediction indicates biometric data in the frames was spoofed, the method includes changing the user head motion type prediction to no motion. The method also includes storing the user head motion type prediction in a buffer and determining a final user head motion type detected for the frames.