摘要
【目的】为了保证多模态医学图像的安全性,提出了一种基于遗传编码和耦合超混沌映射的多模态医学图像保护算法。【方法】在混沌序列生成阶段,提出了CMC-CTBCS耦合模型,以正弦映射(Sine)和无限折叠映射(ICMIC)作为种子映射,构建出三维Sine-ICMIC级联映射(3D-SICM),并将其应用于多模态医学图像加密的设计。图像加密阶段主要包括三部分:多模态医学图像融合算法、基于边缘检测的零水印认证算法、遗传编码选择加密算法。首先,采用基于相位一致性和局部拉普拉斯能量的NSCT域多模态医学图像融合方案,对磁共振成像(MRI)和单光子发射计算机断层成像(SPECT)进行融合,并在融合过程中实现基于边缘检测的零水印信息认证,进而生成融合后的标记图像,然后采用基于动态DNA编码和6比特RNA-S盒替换的选择性加密方案对标记图像进行加密。【结果】理论分析和数值模拟表明,该算法具有良好的安全性。
【Purposes】In order to ensure the security of multimodal medical images,a multimodal medical image protection algorithm based on genetic coding and coupled hyperchaotic mapping is proposed.【Methods】In the stage of chaotic sequence generation,the CMC-CTBCS coupling model was proposed,and the three-dimensional Sine-ICMIC cascade mapping(3D-SICM)was constructed with Sine mapping(Sine)and iterative chaotic map with infinite collapses(ICMIC)as seed mapping,which was applied for the design of multimodal medical image encryption.The stage of image encryption mainly included three parts:multimodal medical image fusion algorithm,zero watermarking authentication algorithm based on edge detection,and genetic coding selection encryption algorithm.First,a multimodal medical image fusion scheme in NSCT domain based on phase consistency and local Laplace energy was used to fuse magnetic resonance imaging and single photon emission computed tomography.Zero watermark information authentication based on edge detection was realized in the fusion process,and the fused mark images were generated.The tagged images were then encrypted by using a selective encryption scheme based on dynamic DNA coding and 6-bit RNAS-box substitution.【Results】Theoretical analyses and numerical simulations show that the algorithm has good security.
作者
常锐云
冯秀芳
张昊
CHANG Ruiyun;FENG Xiufang;ZHANG Hao(College of Software,Taiyuan University of Technology,Jinzhong,Shanxi,China;College of Information and Computer,Taiyuan University of Technology,Jinzhong,Shanxi,China)
出处
《太原理工大学学报》
北大核心
2025年第3期515-523,共9页
Journal of Taiyuan University of Technology
基金
山西省重点研发计划(项目编号:202102020101007)。
作者简介
第一作者:常锐云(1997−),硕士研究生,(E-mail)changruiyun0927@126.com;通信作者:冯秀芳(1966−),博士,教授,主要从事计算机应用、物联网技术、人工智能、无线传感器网络覆盖和定位、虚拟现实技术等,(E-mail)feng_tyut@126.com。
