为了提高多图像加密的安全性, 同时解决多图像加密系统数据量大的问题, 采用了基于分块压缩感知和改进幻方变换的加密方法。加密过程中, 充分利用了混沌序列对初始值的敏感性, 解决基于传统幻方变换的加密算法周期性的问题; 结合分块压缩感知的方法, 减少加密系统的数据量。对4幅256×256的灰度图像进行加密测试。结果表明, 系统加密时间只需要0.98s, 重建图像的质量高, 相关系数值均高于0.99, 峰值信噪比值均大于35dB;该算法在减少加密系统的数据量的同时进一步提高了系统的安全性。该算法实现容易, 能高效安全地完成多图像加密。

In order to improve the security of multi-image encryption and solve the problem of large amount of data in multi-image encryption system, the encryption method based on block compressed sensing and improved magic square transform was adopted. During the encryption process, the sensitivity of chaotic sequence to initial value was fully utilized and the periodicity of encryption algorithm based on traditional magic square transformation was solved. Combining the block compression sensing method, the amount of data in the encryption system was reduced. Four 256×256 gray-scale images were encrypted and tested. The results show that the encryption time of the system is only 0.98s. The quality of the reconstructed image is high. The correlation coefficients were higher than 0.99. Peak signal-to-noise ratio (PSNR) values are greater than 35dB. This algorithm reduces the amount of data in the encryption system and further improves the security of the system. The algorithm is easy to implement and can complete multi-image encryption efficiently and safely.