Research On Image Joint Compression-Encryption Algorithm Based On Improved EZW Coding

With the continuous development of Internet and multimedia technology, multimedia data transmission in network is different from the past, it has become one of the main means for people to exchange data and communicate with each other. Therefore, how to transmit multimedia data securely and effectively has become a significant issue in recent years and brought new challenge to the image compression and encryption.Image compression and encryption are mostly realized separately, and generally image encryption is done after image compression. This kind of method didn’t fully consider the similarity between compression and encryption. So it ends up with a complicated algorithm and consuming more resources. To solve these problems, we proposed an algorithm which can complete image compression and encryption at same time.On the analysis of existing image compression and image encryption algorithm, an image joint compression-encryption algorithm base on improved EZW coding is proposed. The algorithm improves the traditional EZW coding, forms the set by zero tree, and utilizes the correlation between coefficients and sets to encode. LL sub band coefficients are encoded separately because the value of them is large and the correlation is weaker than the other band, this method can restore the image information as much as possible when the rate is fixed. Arithmetic coder is used to encode the decisions of improved EZW coding. The parameters of arithmetic coder are amended by keys to accomplish image joint compression and encryption.Kinds of amendment are proposed which can fully use the relations between sets and coefficients, different sub band can be selected and encrypted by user’s need, so the flexibility of image joint compression and encryption is improved.The simulation results show that the proposed algorithm can realize compression and encryption simultaneously. Taking example for Lena, the PSNR of reconstructed image can achieve more than35dB when the rate is0.5, it shows that the most of image information can be recovered with the low rate. After amending the parameters of arithmetic coder, the PSNR of reconstructed image decrease0.5, it shows that the effect of modification is low. The PSNR of reconstructed image can reduce to9.06dB when the key of decoder is incorrect, so the algorithm can protect the image information effectively.The structure of joint image compression-encryption algorithm is simple. Compared with the other algorithms at the same compression ratio and security requirement, this algorithm is characterized with lower complexity, more flexibility and consuming less resource, so it meets the requirements of the time to implement image compression and encryption at the same time.