A novel 3D vector decomposition for color-image encryption

In this paper, a novel 3D vector decomposition is proposed for color-image cryptosystem, in which a 3D vector is decomposed into two 3D vectors with random size in a random plane for providing a reliable security constraint. The technique of 3D vector decomposition, as far as we know, firstly offers a three-component encryption with one action, which fits well color-image encryption, and outputs a real ciphertext, which is convenient for recording and transmission. Furthermore, we employ an 1D chaos, which has strong chaotic properties, and reality-preserving fractional Hartley transform to cooperate 3D vector decomposition for constructing a color-image cryptosystem. Therefore, the proposed cryptosystem accomplishes improved security by reducing the single-channel attack risk in individual color-image encryption and avoiding the vulnerable channel in sequential color-image encryption as well as reduced the amount of data of transform-based cryptosystem by avoiding the complex output. Also, it has the advantages of strong chaotic performances, large key space and high key sensitivity, which is highly robust against various attacks. Experimental results show the effectiveness and superiority of the proposed cryptosystem.