Path relinking and AES cryptography in color image steganography

Abstract

Each day the Internet use grows up and connects more people, increasing the information flux and the need for transmitting data in a secure way. One of the ways to protect the data sent over the web is to conceal the relevant information inside a typical image, diverting the attention and hiding this data from intruders. This work proposes a hybrid heuristic, combining two methods of computational intelligence: a genetic algorithm and the path relinking refinement, to help solving an image processing issue. It also incorporates the AES advanced cryptography algorithm, to improve the hidden data security. Computational results show that the proposed algorithm outperforms the LSB (least significant bits) substitution technique and also the results of other nine works reported on the literature, concerning the quality of stego image. The inclusion of the path relinking procedure and the possibility to hide data inside colorful images can significantly improve the performance of genetic algorithms in image steganography, increasing the space available for information hiding by more than three times when compared to steganography using grayscale cover images, mostly related to a better least significant bits usage. Anything at all, including documents, text or compressed files, or yet an executable can be hidden inside the cover image, which considerably widens the scope of appliances of this work