SigTem: A Non-Invertible Technique for Online Signature Template Protection

In the digital world, ensuring biometric template security is critical, yet many existing methods prioritize efficiency over protection, leaving sensitive data vulnerable. To address this gap, we propose a robust and efficient approach for securing online signature templates without compromising performance. Our method transforms online signature features into quantized feature vectors encoded as non-invertible binary bitstrings. A user-specific key ensures unique bitstrings per session, effectively preventing attacks. By converting signatures into fixed-length bitstrings, our approach enhances computational efficiency while preserving accuracy. Unlike existing methods, it does not rely on strict signature positioning, ensuring adaptability in real-world scenarios.