一种基于超声波感知手部特征的身份认证方法及系统
By employing active acoustic sensing technology and probabilistic adaptive data augmentation strategies, the hand biometric matrix is extracted and expanded, and dynamically updated using an authentication neural network. This solves the problems of poor biometric consistency and high user registration costs in existing technologies, achieving efficient and secure identity authentication.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING UNIV OF INFORMATION SCI & TECH
- Filing Date
- 2026-04-27
- Publication Date
- 2026-07-17
AI Technical Summary
Existing ultrasonic sensing solutions struggle to handle short-duration, small-amplitude tapping gestures in password input scenarios, resulting in poor biometric consistency, high user registration costs, and a decline in authentication accuracy due to static models failing to adapt to user behavior drift.
Active acoustic sensing technology is used to identify hand reflection signals. A biometric matrix is extracted through audio signal preprocessing. A probabilistic adaptive frame-level data augmentation strategy is used to expand the registered features and dynamically update them through an authentication neural network to construct a high-dimensional feature distribution to improve recognition accuracy and robustness.
It significantly improves the convenience and security of user interaction, reduces registration costs, increases identification accuracy and authentication efficiency, enhances defense against forgery attacks, adapts to complex environments, and ensures the effectiveness of the long-term authentication model.
Smart Images
Figure CN122113074B_ABST