为保证数字图像的安全性,提出了一种压缩图像的三维混沌加密算法。该算法是通过对已压缩的数据流进行加密而实现的。首先采用基于小波的Contourlet变换的类等级树集合分割(SPIHT)编码算法对明文图像进行压缩,得到压缩数据流,然后将压缩数据流映射为一个三维位矩阵;利用Lorenz混沌映射产生混沌序列,并对其进行预处理得到比特值序列,根据比特值序列对上述三维位矩阵进行置乱和替代操作;将置乱和替代后的位矩阵重新映射为数据流,并对其进行解码和反变换操作,得到加密后的压缩图像。实验结果表明,产生的比特值序列具有较好的随机性,加密算法的密钥空间很大,对密钥非常敏感,子密钥和明文有关,能有效抵抗已知明文攻击,结合压缩技术,能有效提高存储和传输效率。

In order to ensure the security of digital image effectively,a 3D chaotic encryption scheme for compressed image is proposed. The scheme is realized by encrypting the datum stream achieved from compressing the original image. Firstly,the wavelet-based contourlet coding using an SPIHT-like algorithm is utilized on the plain image,and as a result,the compressed datum stream could be achieved which is then mapped to a 3D bit matrix. Afterward,a chaotic sequence is generated by Lorenz chaotic map,and it is preprocessed to a bit sequence which is used to permute and substitute the elements of the 3D bit matrix achieved above. In the end,the processed 3D matrix is mapped back to datum stream,and the encrypted compressed image could be achieved if decoding and inverse transform are performed on the datum stream. Experimental results show that the bit sequence generated by the chaotic sequence has a good randomicity and the proposed encryption scheme not only has a large key space but also is very sensitive to the security key. Moreover,its sub-key is relative to the plain image which could effectively resist the known-plaintext attack,and since the algorithm is carried out in the compressed domain it could increase the efficiency of storage and transmission.