Image Cryptosystem for Visually Meaningful Encryption Based on Fractal Graph Generating

Abstract

Image information security is an important research direction in the field of information security. In existed image encryption schemes, original image is transformed to meaningless random noise signal which would draw attention and thus attract attacks. In this work, aiming to encrypt image to visually meaningful encrypted image, a novel image cryptosystem based on fractal graph generating is proposed which is called Visually Meaningful Image Encryption (VMIE) scheme. The VMIE scheme is to encrypt plain images into fractal scenery or fractal plant images by using fractal graph generation parameters as keys. According to these keys, receiver can get the plain image from fractal image by a reverse process. Based on the pixel value distribution characteristics of different plain images, different fractal transformations can be selected to produce different fractal images, which can better camouflage the plain images and make the encrypted plain images safer. In this paper, a new image encryption method is proposed, which is different from VMIE based on information hiding with limited hiding capacity, and from VMIE based on image sharing, which takes up a large amount of storage space and image size. Simulation experiments show that the proposed new scheme has low computation complexity and high security and the encrypted plain images have well understandability, strong resistance to compression and small data increase.