Nonlinear image authentication algorithm based on double fractional Mellin domain

In this paper, we propose a novel dual-user image authentication algorithm based on the double fractional Mellin transform. We perform a security analysis of the nonlinear cryptosystem based on the fractional Mellin transform and show its vulnerability. In the proposed algorithm, polar decomposition and sparse multiplexing are additionally applied to generate a ciphertext. During the encryption process, polar decomposition generates two private keys that can be utilized on the dual-user authentication platform. The proposed scheme has a large key space and is robust against several attacks such as contamination attacks (noise and occlusion), brute force attacks, plain-text attacks, and special iterative attacks. In addition, we carry out a comparison with a similar existing scheme for the proposed algorithm. Simulated results indicate that the proposed authentication algorithm is feasible and robust.