Image Encryption Algorithms Based On Arithmetic Coding

In recent years, inspired by the fast development of computer technologies and multimedia technologies, digital images and videos have been widely transmitted on networks. In the mean time, the security and privacy information contained in these multimedia data have attracted wide attention of people. Cryptography, as a key technology to guarantee the information security of transmitted data, is facing new challenges in the application of multimedia transmission. First, different from text data, the size of multimedia data is very large and multimedia data have very high degree of redundancy. So if we use traditional cryptography on these multimedia data, it will consume considerable time and computing resources. Second, the encrypted data often grows very large in size, which makes it not suitable for transmission, sometimes even unsustainable by the transmission channel. In order to solve the aforementioned problems, it is necessary to combine both encryption and compression. In this way, we can not only guarantee the information security and image quality, but also needs less resource for storing, recording and transmitting these multimedia data to meet the requirements of economical efficiency, reliability and high quality. Since digital images are the most common used multimedia data, and the format of most other multimedia data are designed based on digital images, we, in this paper, focus on the compression and cryptography of digital images. Considering that the effectiveness of compression by using arithmetic coding has been widely agreed, and it has been adopted in the updated standards of multimedia data compression such as JPEG2000 and H.264, in this paper, we mainly concentrate on the application and improvement of arithmetic coding in the compression and encryption of digital images.The main contents of this paper are listed as follows.①Arithmetic coding theory and its advantage are introduced. The arithmetic coding theory is described in details including the definition of arithmetic coding, coding model, coding process and the properties of arithmetic coding, and we also analyze why it is suitable for compression and encryption.②The random arithmetic coding (RAC) algorithm proposed by Grangetto et al. is improved effectively. Firstly, a generalized description of RAC is given. Secondly, the weaknesses of this algorithm are presented. Based on these analyses, a secure arithmetic coding scheme is proposed, and some tests about security and efficiency are carried out.③A novel secure arithmetic coding with excellent compression efficiency is proposed, which is based on Markov model, and have been applied it to the image encryption. It makes the arithmetic coding secure and have a better compression efficiency than RAC by using the higher-orders model, and the rule of permuting probability intervals, which ensure the image transmit in the internet securely and efficiently.④Due to all the cryptosystem based on arithmetic coding being vulnerable to chosen plaintext attack, which is the security holes of the algorithm based on Markov model, an improved scheme against the chosen plaintext attack has been proposed. Experimental results and security analyses indicate that, the algorithm can not only resist to existing attacks based on arithmetic coding, but also be immune to other cryptanalysis.Finally, the conclusion is summed up. Some deficiencies as well as further work are also given.