基于多混沌和分数Fourier的光学图像加密算法

Optical image encryption algorithm based on multi-chaos and fractional Fourier

针对双随机相位编码光学图像加密系统的非线性不足和密钥空间小的问题,提出一种基于多混沌和分数Fourier变换的光学图像加密算法.首先,迭代分数阶Chen混沌系统生成三组混沌序列,分别置乱明文图像的三基色分量以减小图像像素相关性.然后,利用量子细胞神经网络超混沌系统调制随机相位模板,以其复杂的非线性动力学特征弥补双随机相位编码系统非线性不足的缺陷.其次,通过使用分数阶混沌、超混沌系统和二维分数Fourier变换使加密算法的密钥空间达到了2^(765).最后,密钥敏感性测试、相关性分析、已知明文攻击、噪声攻击和剪切攻击等实验表明本算法具有密钥敏感性强、密文图像像素相关性低和抗攻击性强的优点.

To deal with the problem of lacking nonlinearity and the low-key space in the Double Random Phase Encoding(DRPE)system,an optical image encryption algorithm based on multiple chaotic systems and the fractional Fourier transform is proposed. Firstly,three groups of chaotic sequences are generated through the iteration of the fractional chaotic system to scramble each of the three primary colors respectively,so that the correlation of the pixels in the explicit image is reduced. Meanwhile,the random phase mask is modulated in the DRPE module by using the hyperchaotic system of quantum cellular neural network,whose complicated nonlinear dynamic features make up for the shortcomings of the DRPE system in this regard. Next,the key space is 2^765 by using fractional chaos,hyper-chaotic and two-dimensional fractional Fourier transform. Finally,by analyzing the results of experiments,such as key sensitivity tests,correlation analysis,known plaintext attack,noise attack and occlusion attack,it can be concluded that the proposed algorithm has some advantages of strong key sensitivity,low pixel correlation of ciphertext image and high-resistance against attacks.