Multibiometric Template Security Using CS Theory – SVD Based Fragile Watermarking Technique

Protection of biometric template against spoofing or modification attack at system database is major issue in multibiometric system. Hence fragile digital watermarking technique is one of the solutions for biometric template protection against these attacks. In this paper, fingerprint watermarking technique based on SVD and Compressive Sensing theory proposed for protection of biometric template at system database of multibiometric system. This technique does not embed the fingerprint directly into face image instead using the concept of compressive sensing fingerprint convert into sparse measurements. The sparse measurements is generated at time of fingerprint embedding process and extracted from watermarked face image at extraction side for reconstruction of original watermark fingerprint image. SSIM value between original watermark fingerprint image and reconstructed watermark fingerprint image is the deciding factor for cross verification of individual. The experimental results show that the proposed technique does not affect verification and authentication performance of multibiometric system. Key-Words: Compressive Sensing Theory, Fragile, Multibiometric, SVD, Template Protection, Watermarking