改进Arnold算法和超混沌系统的医学图像加密研究

Research on medical image encryption algorithm based on modified Arnold transformation and hyper chaos

针对医学图像的信息安全问题,提出了一种基于改进Arnold置乱和超混沌系统的医学图像加密算法。利用Lorenz超混沌系统生成四组密钥序列;通过最低有效位置换,实现对病人信息的隐藏;利用改进Arnold置乱算法,结合像素位随机取反和异或实现对图像信息的加密。最后分析了算法的安全性。实验结果表明:算法密钥空间为1067数量级,算法安全性较高,并解决了DICOM(Digital Imaging and Communications in Medicine)图像中病人信息的加密问题,实现了对图像数据和病人信息数据的双重保护,可适用于不同位深,不同源,不等长宽的DICOM格式医学图像的加密,保护病人隐私。

According to the security problems of patient information and image data in medical images, the paper proposed an encryption algorithm based on the modified Arnold transformation and hyper chaos. The algorithm uses Lorenz maps to build four key sequences according to some rule; uses Least Significant Bit replacing method to hide the patient information; and combined with Pixels' bit-reversed randomized method and XOR method, the modified Arnold transformation is used to encrypt the image data. The security of the algorithm is analyzed. Experimental results show that the algorithm achieves a larger key space in the order of 1067 magnitude, and has a good security. The algorithm can solve the issue of encrypting patient information in DICOM medical digital images. The algorithm, which has the complex rules to generate key sequences, encrypts both image data and patient data. It is suitable for different types of DICOM images, and protects patient privacy.