Abstract: In order to solve the defects as low security and decryption image distortion in current multi-image encryption technology, a new multi-image synchronous lossless encryption algorithm based on hyper-complex fusion model and fractional order chaotic dynamics permutation is proposed. The fractional order logistic map was constructed, and the dynamical permutation mechanism was designed by considering the pixels of each plain to permute all the plains. And four DCT coefficient matrixes were got based on discrete cosine transformation. Then the hyper-complex fusion model was designed to lossless fuse the four coefficient matrixes for obtaining composite matrix. Then the complex scrambling image was got by inverse DCT transform. Finally, chaotic mask was got by iterating the fractional order chaotic map again, and pixel encryption mechanism was designed by jointing the FrFT transform to realize the Multi-image encryption. The experimental data show that this algorithm has higher security and restoration quality. 

Key words: Multi image encryption, hyper complex fusion model, fractional order chaotic map, dynamic permutation mechanism, chaotic phase mask, distortion

摘要： 为了解决当前多图像加密技术存在的的安全性不高、解密图像存在失真的不足，设计了基于超复数融合模型与分数阶混沌动态混淆的多图像同步无损加密算法。构建分阶Logistic混沌映射，考虑每幅明文的像素，设计了动态混淆机制，对每幅明文进行置乱；基于(Discrete Cosine Transform，DCT)机制，输出4个置乱图像的DCT系数矩阵；设计超复数融合模型，将4个DCT系数矩阵进行无损融合，获取复合矩阵；利用逆DCT变换，输出复合置乱图像。再次迭代分阶Logistic混沌映射，获取混沌掩码，并联合( Fractional Fourier Transform，FrFT)变换，设计像素加密机制，实现多图像同步加密。试验数据显示，与当前多图像加密技术相比，所提同步加密算法具有更高的安全性与复原质量。

关键词: 多图像加密, 超复数融合模型, 分阶Logistic混沌映射, 动态混淆机制, 混沌相位掩码, 失真度