Discrete wavelet transform and compressive sensing based multibiometric watermarking — A novel approach to embed watermark into biometrie

Multibiometric system is used two or more than two characteristics of same individual for identification and authentication. This is improved security and authenticity of individual. The issued regarding multibiometric system is template protection against modification attack at system database and at communication channel between two modules of system. This paper proposed a novel multibiometric watermarking technique in wavelet domain using compressive sensing theory framework. The main idea of proposed technique is embed sparse measurements of watermark biometric image which is generated using details wavelet coefficients of watermark biometric image and random seed which is embedding into approximation wavelet coefficients of host biometric image of same individual. This proposed technique is generated multibiometric template which provides security against modification attack over communication channel and at system database because of secure biometric information is embed into another biometric of same individual. Compressive sensing theory framework is used to provide more computation security and dimensional reduction to watermark biometric image before embedding into host biometric image. This proposed watermarking technique has robust against all possible watermarking attacks.