Chaos-based model for encryption and decryption of digital images

Abstract

This paper introduces a secure chaos-based model for ciphering and deciphering of digital images. The proposed approach is composed of successive confusion and diffusion stages. The confusion stage is repeated n rounds using a different key in each round. The output of the confusion stage is subjected to diffusion stage which is repeated m rounds with a different key for each round. The nested iterations in the confusion and diffusion stages with a different key for each round enlarges the key space which enhances the proposed image cryptosystem security level. A security investigation is done on a family of 2D chaotic confusion maps to select the one with highest security level to be used with the proposed image cryptosystem. The results demonstrated that the Standard map has the highest security level among the examined 2D chaotic confusion maps because it is more complicated and it has a large key space. The proposed image cryptosystem is compared to other three recent image cryptosystems using different security analysis factors including statistical tests, key space analysis, information entropy test, maximum deviation analysis, irregular deviation analysis, and avalanche effect differential analysis. The results demonstrated that, the proposed image cryptosystem with Standard map outperforms all of the other examined image encryption techniques from security point of view.