SILENT STAMP – UNIQUE ID CREATION FOR BIOMETRIC VERIFICATION

The project demonstrates the potential of zero-bit watermarking as a powerful solution for biometric security. By embedding fingerprint features into iris images and generating a secure encrypted master share, the proposed system overcomes the limitations of conventional watermarking techniques that often cause distortion or lack robustness.   The main advantage of this approach is its distortion-free nature, ensuring that biometric images retain their original quality while being secured against unauthorized access. The use of BER and PSNR metrics provides objective evidence of system reliability and performance. Furthermore, by combining two biometric modalities (iris and fingerprint), the system enhances uniqueness and reduces the chances of duplication or identity forgery.   With potential applications spanning government identification systems, banking, healthcare, military, and corporate sectors, the proposed framework provides a scalable and practical solution for modern authentication systems. The project not only addresses current research gaps but also sets the foundation for future advancements in biometric watermarking, where security, accuracy, and robustness are critical.

[1]         Preetam Am Rit and Naman Baranwal, “ConvNet-HIDE: Deep-Learning-Based Dual Watermarking,” IEEE Computer Society, 2024. [2]         Mahendra Rathor and Aditya Anshul, “Securing Reusable IP Cores Using Voice Biometric Based Watermark,” IEEE Computer Society, 2024. [3]         Shuo Shao, Wenyuan Yang, Hanlin Gu, Zhan Qin, Lixin Fan, Qiang Yang, and Kui Ren, “FedTracker: Ownership Verification & Traceability,” IEEE Computer Society, 2025. [4]         Ziting Ren, Yucong Duan, Qi Qi, and Lanhua Luo, “Data-Free Fingerprinting Technology for Biometric Classifiers,” Elsevier, 2025. [5]         Wenzhong Shen and Ji Rong, “DNN-Based Iris Watermarking for Biometric Security,” Peer-Reviewed Journal, 2022.